2_DC_GraphModel.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "961764cd",
   "metadata": {},
   "source": [
    "# DeepChem\n",
    "\n",
    "- [Installation](https://github.com/deepchem/deepchem#installation)\n",
    "- [Tutorial](https://deepchem.readthedocs.io/en/latest/get_started/tutorials.html)\n",
    "- [Sample Notebooks](https://github.com/deepchem/deepchem/tree/master/examples/tutorials) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ee150959",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "\n",
    "import deepchem as dc\n",
    "\n",
    "from sklearn.metrics import r2_score, mean_squared_error"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d6608e5",
   "metadata": {},
   "source": [
    "# Parameters , Settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3a4e6075",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_filepath = './data/ESOL.csv'   \n",
    "\n",
    "model_dir = './result'   # folder to save fitted model\n",
    "save_hyperparms = './result/Hyperparameter'   # folder to save models during hyperparameter optimization\n",
    "\n",
    "n_tasks = 1   # No. of tasks (No. of dependent variables)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "46d70723",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialze the metrics\n",
    "# https://deepchem.readthedocs.io/en/latest/api_reference/metrics.html\n",
    "metric_r2 = dc.metrics.Metric(dc.metrics.r2_score)\n",
    "metric_mse = dc.metrics.Metric(dc.metrics.mean_squared_error)\n",
    "metrics = [metric_r2, metric_mse]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef8fb357",
   "metadata": {},
   "source": [
    "# Load Data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "9c08bab7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1128, 2)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(data_filepath)   \n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "1b354f8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "smiles = data['smiles']   # should be 1D \n",
    "y = data['measured log solubility in mols per litre']   # can be 1D or 2D"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44014bc3",
   "metadata": {},
   "source": [
    "# Make Dataset for DeepChem"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "288a5cf4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "deepchem.feat.mol_graphs.ConvMol"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "''' ConvMolFeaturizer '''\n",
    "# Duvenaud graph convolutions ; Can be used with Keras models\n",
    "# https://deepchem.readthedocs.io/en/latest/api_reference/featurizers.html#convmolfeaturizer\n",
    "\n",
    "featurizer = dc.feat.ConvMolFeaturizer()\n",
    "features = featurizer.featurize(smiles)   # numpy array, it returns a graph object for every molecule\n",
    "type(features[0])   # deepchem.feat.mol_graphs.ConvMol"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6e1fff2c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<NumpyDataset X.shape: (1128,), y.shape: (1128,), w.shape: (1128,), task_names: [0]>\n"
     ]
    }
   ],
   "source": [
    "# \"NumpyDataset\" class stores datasets in memory. This works fine for smaller datasets, but is less convenient for larger datasets. \n",
    "# For large datasets, use \"DiskDataset\" which writes dataset to a folder. \n",
    "# https://deepchem.readthedocs.io/en/latest/api_reference/data.html\n",
    "# https://github.com/deepchem/deepchem/blob/master/examples/tutorials/02_Working_With_Datasets.ipynb\n",
    "dataset = dc.data.NumpyDataset(X=features, y=np.array(y))\n",
    "print(dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afd72238",
   "metadata": {},
   "source": [
    "# Split Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e07dd970",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<NumpyDataset X.shape: (226,), y.shape: (226,), w.shape: (226,), ids: [148 1014 1021 ... 835 559 684], task_names: [0]>\n"
     ]
    }
   ],
   "source": [
    "# \"random splitting\" sometimes overestimates model’s performance, especially for small data or imbalance data. \n",
    "# The dc.splits provides more methods and algorithms to evaluate the model’s performance appropriately, like cross validation or splitting using molecular scaffolds.\n",
    "# https://deepchem.readthedocs.io/en/latest/api_reference/splitters.html\n",
    "\n",
    "splitter = dc.splits.RandomSplitter()\n",
    "\n",
    "train_dataset, test_dataset = splitter.train_test_split(dataset=dataset, frac_train=0.8, seed=0)\n",
    "print(test_dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad0dbba0",
   "metadata": {},
   "source": [
    "# Model , Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b808c292",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_14:0\", shape=(353,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_13:0\", shape=(353, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_17:0\", shape=(1314,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_16:0\", shape=(1314, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_20:0\", shape=(1149,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_19:0\", shape=(1149, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_23:0\", shape=(132,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_22:0\", shape=(132, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_11:0\", shape=(353,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_10:0\", shape=(353, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_13:0\", shape=(1314,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_12:0\", shape=(1314, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_15:0\", shape=(1149,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_14:0\", shape=(1149, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_17:0\", shape=(132,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_16:0\", shape=(132, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_14:0\", shape=(353,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_13:0\", shape=(353, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_17:0\", shape=(1314,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_16:0\", shape=(1314, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_20:0\", shape=(1149,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_19:0\", shape=(1149, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_23:0\", shape=(132,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_22:0\", shape=(132, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_14:0\", shape=(349,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_13:0\", shape=(349, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_17:0\", shape=(1264,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_16:0\", shape=(1264, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_20:0\", shape=(1131,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_19:0\", shape=(1131, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_23:0\", shape=(140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_22:0\", shape=(140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_11:0\", shape=(349,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_10:0\", shape=(349, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_13:0\", shape=(1264,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_12:0\", shape=(1264, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_15:0\", shape=(1131,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_14:0\", shape=(1131, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_17:0\", shape=(140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_16:0\", shape=(140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_14:0\", shape=(349,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_13:0\", shape=(349, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_17:0\", shape=(1264,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_16:0\", shape=(1264, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_20:0\", shape=(1131,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_19:0\", shape=(1131, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_23:0\", shape=(140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_22:0\", shape=(140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    }
   ],
   "source": [
    "# https://deepchem.readthedocs.io/en/latest/api_reference/models.html#graphconvmodel\n",
    "# This Class uses Keras models\n",
    "model = dc.models.GraphConvModel(n_tasks=n_tasks,              # No. of tasks \n",
    "                                 graph_conv_layers=[64, 64],   # Width of channels for the Graph Convolution Layers\n",
    "                                 dense_layer_size=128,         # Width of channels for Atom Level Dense Layer before GraphPool\n",
    "                                 dropout=0.2,                  # Dropout probablity to use for each layer. The length of this list should equal len(graph_conv_layers)+1 (one value for each convolution layer, and one for the dense layer). Alternatively this may be a single value instead of a list, in which case the same value is used for every layer.\n",
    "                                 mode='regression',            # Either “classification” or “regression”\n",
    "                                 batch_size=100,               # Batch size for training and evaluating\n",
    "                                 model_dir=model_dir,          # Directory on disk where the model will be stored. If this is None, a temporary directory is created.\n",
    "                                 learning_rate=0.001\n",
    "                                 )\n",
    "\n",
    "# https://deepchem.readthedocs.io/en/latest/api_reference/models.html#kerasmodel\n",
    "loss_avg = model.fit(train_dataset, nb_epoch=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "30adbf9c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"private__graph_conv_keras_model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "graph_conv (GraphConv)       multiple                  102144    \n",
      "_________________________________________________________________\n",
      "graph_conv_1 (GraphConv)     multiple                  87360     \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo multiple                  256       \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch multiple                  256       \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch multiple                  512       \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            multiple                  0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          multiple                  0         \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          multiple                  0         \n",
      "_________________________________________________________________\n",
      "graph_pool (GraphPool)       multiple                  0         \n",
      "_________________________________________________________________\n",
      "graph_pool_1 (GraphPool)     multiple                  0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                multiple                  8320      \n",
      "_________________________________________________________________\n",
      "graph_gather (GraphGather)   multiple                  0         \n",
      "_________________________________________________________________\n",
      "trim_graph_output (TrimGraph multiple                  0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              multiple                  257       \n",
      "=================================================================\n",
      "Total params: 199,105\n",
      "Trainable params: 198,593\n",
      "Non-trainable params: 512\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.model.summary()   # must be after \"model.fit\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f25953c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: {'r2_score': 0.7225232445921097, 'mean_squared_error': 1.224374317157278}\n",
      "Test set score: {'r2_score': 0.6660339328231306, 'mean_squared_error': 1.4380285752099478}\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model\n",
    "print(\"Training set score:\", model.evaluate(train_dataset, metrics))\n",
    "print(\"Test set score:\", model.evaluate(test_dataset, metrics))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6b8dcfeb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['./result/ckpt-1']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.get_checkpoints()   # location of fitted model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec0a6f18",
   "metadata": {},
   "source": [
    "### Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ddd53f7d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# input to the \"predict\" is \"dataset\"\n",
    "y_pred = model.predict(test_dataset)   # numpy array [n, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "7ec61778",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test dataset (R2) :  0.6660339328231306\n",
      "Test dataset (RMSE) :  1.1991782916689027\n"
     ]
    }
   ],
   "source": [
    "# Since we have predicted values, we can use scikit-learn methods to calculate metrics as usual.\n",
    "y_test = test_dataset.y\n",
    "\n",
    "print(\"Test dataset (R2) : \", r2_score(y_test, y_pred))\n",
    "print(\"Test dataset (RMSE) : \", np.sqrt(mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c155996e",
   "metadata": {},
   "source": [
    "# Reload Saved Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "eb504428",
   "metadata": {},
   "outputs": [],
   "source": [
    "del model\n",
    "\n",
    "# Initialize a model with the same parameters when you defined model for the 1st time (before training)\n",
    "model = dc.models.GraphConvModel(n_tasks=n_tasks,              # No. of tasks \n",
    "                                 graph_conv_layers=[64, 64],   # Width of channels for the Graph Convolution Layers\n",
    "                                 dense_layer_size=128,         # Width of channels for Atom Level Dense Layer before GraphPool\n",
    "                                 dropout=0.2,                  # Dropout probablity to use for each layer. The length of this list should equal len(graph_conv_layers)+1 (one value for each convolution layer, and one for the dense layer). Alternatively this may be a single value instead of a list, in which case the same value is used for every layer.\n",
    "                                 mode='regression',            # Either “classification” or “regression”\n",
    "                                 batch_size=100,               # Batch size for training and evaluating\n",
    "                                 model_dir=model_dir,          # Directory on disk where the model will be stored. If this is None, a temporary directory is created.\n",
    "                                 learning_rate=0.001\n",
    "                                 )\n",
    "\n",
    "# fitted model is saved in \"model_dir\" and we can restore it.\n",
    "model.restore()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "5c5c5588",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test dataset (R2) :  0.6660339328231306\n",
      "Test dataset (RMSE) :  1.1991782916689027\n"
     ]
    }
   ],
   "source": [
    "# Test prediction\n",
    "# input to the \"predict\" is \"dataset\"\n",
    "y_pred = model.predict(test_dataset)   # numpy array [n, 1]\n",
    "\n",
    "print(\"Test dataset (R2) : \", r2_score(y_test, y_pred))\n",
    "print(\"Test dataset (RMSE) : \", np.sqrt(mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83a36d13",
   "metadata": {},
   "source": [
    "# Uncertainty\n",
    "\n",
    "Uncertainties are estimates, not rigorous measurements. Most of them are pretty good, but you should not put too much confidence in any single value.\n",
    "\n",
    "It involves repeating the prediction many times with different dropout masks.\n",
    "\n",
    "If a model supports uncertainty, it MUST use dropout on every layer, and dropout most be enabled during uncertainty prediction. Otherwise, the uncertainties it computes will be inaccurate\n",
    "\n",
    "https://github.com/deepchem/deepchem/blob/master/examples/tutorials/25_Uncertainty_In_Deep_Learning.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "29e17a8d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.iter\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.beta_1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.beta_2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.decay\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.learning_rate\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.regression_dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.regression_dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel19\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.kernel20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.1.bias20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.0.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.0.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.1.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.1.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.2.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.2.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.regression_dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.regression_dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias15\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.kernel20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.1.bias20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.0.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.0.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.1.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.1.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.2.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.2.beta\n",
      "WARNING:tensorflow:A checkpoint was restored (e.g. tf.train.Checkpoint.restore or tf.keras.Model.load_weights) but not all checkpointed values were used. See above for specific issues. Use expect_partial() on the load status object, e.g. tf.train.Checkpoint.restore(...).expect_partial(), to silence these warnings, or use assert_consumed() to make the check explicit. See https://www.tensorflow.org/guide/checkpoint#loading_mechanics for details.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_14:0\", shape=(342,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_13:0\", shape=(342, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_17:0\", shape=(1212,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_16:0\", shape=(1212, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_20:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_19:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_23:0\", shape=(140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_22:0\", shape=(140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_11:0\", shape=(342,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_10:0\", shape=(342, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_13:0\", shape=(1212,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_12:0\", shape=(1212, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_15:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_14:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_17:0\", shape=(140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_16:0\", shape=(140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_14:0\", shape=(342,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_13:0\", shape=(342, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_17:0\", shape=(1212,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_16:0\", shape=(1212, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_20:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_19:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_23:0\", shape=(140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_22:0\", shape=(140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_14:0\", shape=(366,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_13:0\", shape=(366, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_17:0\", shape=(1260,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_16:0\", shape=(1260, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_11:0\", shape=(366,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_10:0\", shape=(366, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_13:0\", shape=(1260,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_12:0\", shape=(1260, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_17:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_16:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_14:0\", shape=(366,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_13:0\", shape=(366, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_17:0\", shape=(1260,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_16:0\", shape=(1260, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    }
   ],
   "source": [
    "del model\n",
    "\n",
    "# Initialize model with parameter \"uncertainty=True\"\n",
    "model = dc.models.GraphConvModel(n_tasks=n_tasks,              # No. of tasks \n",
    "                                 graph_conv_layers=[64, 64],   # Width of channels for the Graph Convolution Layers\n",
    "                                 dense_layer_size=128,         # Width of channels for Atom Level Dense Layer before GraphPool\n",
    "                                 dropout=0.2,                  # Dropout probablity to use for each layer. The length of this list should equal len(graph_conv_layers)+1 (one value for each convolution layer, and one for the dense layer). Alternatively this may be a single value instead of a list, in which case the same value is used for every layer.\n",
    "                                 mode='regression',            # Either “classification” or “regression”\n",
    "                                 batch_size=100,               # Batch size for training and evaluating\n",
    "                                 model_dir=model_dir,          # Directory on disk where the model will be stored. If this is None, a temporary directory is created.\n",
    "                                 learning_rate=0.001,\n",
    "                                 uncertainty=True\n",
    "                                 )\n",
    "\n",
    "loss_avg = model.fit(train_dataset, nb_epoch=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7983ef86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2.187775 ],\n",
       "       [2.1478465],\n",
       "       [1.9598346],\n",
       "       [1.6264415],\n",
       "       [1.8399079]], dtype=float32)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# https://deepchem.readthedocs.io/en/latest/api_reference/models.html#deepchem.models.KerasModel.predict_uncertainty\n",
    "y_pred, y_std = model.predict_uncertainty(test_dataset)   # numpy array\n",
    "y_std[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05299dca",
   "metadata": {},
   "source": [
    "# Hyperparameter Optimization (Grid Search)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b8aa358f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# First, we split dataset again into train / validation / test. \n",
    "# Hyperparameter Optimization requires validation dataset. Best model is selected based on the metric of validation dataset. \n",
    "splitter = dc.splits.RandomSplitter()\n",
    "train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split(dataset=dataset, \n",
    "            frac_train=0.7, frac_valid=0.15, frac_test=0.15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "6112f56b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Values in \"params_dict\" must be List\n",
    "params_dict = {\n",
    "    'n_tasks': [n_tasks],\n",
    "    'mode': ['regression'],   \n",
    "    'graph_conv_layers': [[64], [64, 64]],\n",
    "    'dropout': [0.2, 0.5]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "a67088e4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_14:0\", shape=(362,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_13:0\", shape=(362, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_17:0\", shape=(1096,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_16:0\", shape=(1096, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_20:0\", shape=(930,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_19:0\", shape=(930, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_23:0\", shape=(180,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_22:0\", shape=(180, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_14:0\", shape=(336,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_13:0\", shape=(336, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_17:0\", shape=(1286,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_16:0\", shape=(1286, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_20:0\", shape=(1152,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_19:0\", shape=(1152, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_23:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_22:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_14:0\", shape=(332,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_13:0\", shape=(332, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_17:0\", shape=(1204,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_16:0\", shape=(1204, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_20:0\", shape=(1056,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_19:0\", shape=(1056, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_23:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_22:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_14:0\", shape=(358,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_13:0\", shape=(358, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_17:0\", shape=(1296,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_16:0\", shape=(1296, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_20:0\", shape=(1152,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_19:0\", shape=(1152, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_23:0\", shape=(152,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_22:0\", shape=(152, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_7/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_14:0\", shape=(323,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_13:0\", shape=(323, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_17:0\", shape=(1182,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_16:0\", shape=(1182, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_20:0\", shape=(945,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_19:0\", shape=(945, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_23:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_22:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_11:0\", shape=(323,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_10:0\", shape=(323, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_13:0\", shape=(1182,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_12:0\", shape=(1182, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_15:0\", shape=(945,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_14:0\", shape=(945, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_17:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_16:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_14:0\", shape=(323,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_13:0\", shape=(323, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_17:0\", shape=(1182,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_16:0\", shape=(1182, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_20:0\", shape=(945,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_19:0\", shape=(945, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_23:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_22:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_14:0\", shape=(354,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_13:0\", shape=(354, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_17:0\", shape=(1254,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_16:0\", shape=(1254, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_20:0\", shape=(1140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_19:0\", shape=(1140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_11:0\", shape=(354,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_10:0\", shape=(354, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_13:0\", shape=(1254,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_12:0\", shape=(1254, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_15:0\", shape=(1140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_14:0\", shape=(1140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_14:0\", shape=(354,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_13:0\", shape=(354, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_17:0\", shape=(1254,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_16:0\", shape=(1254, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_20:0\", shape=(1140,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_19:0\", shape=(1140, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_9/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_9/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_8/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_14:0\", shape=(316,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_13:0\", shape=(316, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_17:0\", shape=(1128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_16:0\", shape=(1128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_20:0\", shape=(1044,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_19:0\", shape=(1044, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_11:0\", shape=(316,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_10:0\", shape=(316, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_13:0\", shape=(1128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_12:0\", shape=(1128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_15:0\", shape=(1044,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_14:0\", shape=(1044, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_17:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_16:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_14:0\", shape=(316,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_13:0\", shape=(316, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_17:0\", shape=(1128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_16:0\", shape=(1128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_20:0\", shape=(1044,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_19:0\", shape=(1044, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_14:0\", shape=(370,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_13:0\", shape=(370, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_17:0\", shape=(1214,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_16:0\", shape=(1214, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_20:0\", shape=(1122,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_19:0\", shape=(1122, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_23:0\", shape=(192,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_22:0\", shape=(192, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_11:0\", shape=(370,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_10:0\", shape=(370, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_13:0\", shape=(1214,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_12:0\", shape=(1214, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_15:0\", shape=(1122,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_14:0\", shape=(1122, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_17:0\", shape=(192,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_16:0\", shape=(192, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_14:0\", shape=(370,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_13:0\", shape=(370, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_17:0\", shape=(1214,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_16:0\", shape=(1214, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_20:0\", shape=(1122,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_19:0\", shape=(1122, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_23:0\", shape=(192,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_22:0\", shape=(192, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_11/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_10/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    }
   ],
   "source": [
    "# Grid Search\n",
    "# https://deepchem.readthedocs.io/en/latest/api_reference/hyper.html#grid-hyperparameter-optimization\n",
    "optimizer = dc.hyper.GridHyperparamOpt(dc.models.GraphConvModel)\n",
    "\n",
    "best_model, best_hyperparams, all_results = optimizer.hyperparam_search(\n",
    "            params_dict,       # dictionary that maps hyperparameter names (strings) to lists of possible parameter values\n",
    "            train_dataset, \n",
    "            valid_dataset,     # dataset used for validation (optimization on valid scores) \n",
    "            metric_r2,         # metric used for evaluation\n",
    "            nb_epoch=50,\n",
    "            use_max=True,      # If True, return the model with the highest score. Else return model with the minimum score.\n",
    "            logdir=save_hyperparms   # The directory in which to store created models. If not set, will use a temporary directory.\n",
    "            )\n",
    "# all_results contains value of metric for validation dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8eb68e61",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 'regression', [64], 0.2)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_hyperparams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "0b689a2f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'_dropouts_0.200000_graph_conv_layers[64]_moderegression_n_tasks_1': 0.746424775042789,\n",
       " '_dropouts_0.500000_graph_conv_layers[64]_moderegression_n_tasks_1': 0.7007858304236246,\n",
       " '_dropouts_0.200000_graph_conv_layers[64, 64]_moderegression_n_tasks_1': 0.6978143017856612,\n",
       " '_dropouts_0.500000_graph_conv_layers[64, 64]_moderegression_n_tasks_1': 0.7285161522316366}"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "a8634424",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['./result/Hyperparameter/0/ckpt-1']"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model.get_checkpoints()   # filename of best model saved in the folder \"save_hyperparms\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d37c66a2",
   "metadata": {},
   "source": [
    "### Load saved best model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "f6933d66",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.iter\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.beta_1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.beta_2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.decay\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer.learning_rate\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.regression_dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.regression_dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.kernel20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.graph_convs.0.bias20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.0.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.0.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.1.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'm' for (root).model.batch_norms.1.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.regression_dense.kernel\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.regression_dense.bias\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel11\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.kernel20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias0\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias1\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias2\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias3\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias4\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias5\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias6\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias7\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias8\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias9\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias10\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias11\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias12\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias13\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias14\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias15\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias16\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias17\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias18\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias19\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.graph_convs.0.bias20\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.0.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.0.beta\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.1.gamma\n",
      "WARNING:tensorflow:Unresolved object in checkpoint: (root).optimizer's state 'v' for (root).model.batch_norms.1.beta\n",
      "WARNING:tensorflow:A checkpoint was restored (e.g. tf.train.Checkpoint.restore or tf.keras.Model.load_weights) but not all checkpointed values were used. See above for specific issues. Use expect_partial() on the load status object, e.g. tf.train.Checkpoint.restore(...).expect_partial(), to silence these warnings, or use assert_consumed() to make the check explicit. See https://www.tensorflow.org/guide/checkpoint#loading_mechanics for details.\n"
     ]
    }
   ],
   "source": [
    "best_model_checkpoints = best_model.get_checkpoints()[0]\n",
    "best_model_folder = os.path.dirname(best_model_checkpoints)   # name of folder where best model is saved\n",
    "\n",
    "del model\n",
    "\n",
    "# Initialize model with best hyperparameters\n",
    "model = dc.models.GraphConvModel(n_tasks=1, mode='regression', graph_conv_layers=[64], \n",
    "                                 dropout=0.2, model_dir=best_model_folder)\n",
    "\n",
    "model.restore()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "754149a8",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: {'r2_score': 0.9459974181683233}\n",
      "Validation set score: {'r2_score': 0.746424775042789}\n",
      "Test set score: {'r2_score': 0.7605521890903217}\n"
     ]
    }
   ],
   "source": [
    "print(\"Training set score:\", model.evaluate(train_dataset, metric_r2))\n",
    "print(\"Validation set score:\", model.evaluate(valid_dataset, metric_r2))\n",
    "print(\"Test set score:\", model.evaluate(test_dataset, metric_r2))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1007209",
   "metadata": {},
   "source": [
    "# Hyperparameter Optimization (Gaussian Process)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5fb3e33",
   "metadata": {},
   "source": [
    "#### Gaussian Process Global Optimization (GPGO)\n",
    "\n",
    "https://deepchem.readthedocs.io/en/latest/api_reference/hyper.html#gaussian-process-hyperparameter-optimization\n",
    "\n",
    "https://github.com/deepchem/deepchem/tree/master/examples/hyperparam_opt\n",
    "\n",
    "It uses pyGPGO to optimize hyperparameters : pip install pyGPGO\n",
    "\n",
    "\n",
    "\"params_dict\" has a different semantics than for GridHyperparamOpt. param_dict[key] must be an int/float, which is used as the center of a search with radius \"search_range\" since pyGPGO can only optimize numerical hyperparameters.\n",
    "\n",
    "\"search_range\" specifies the range of parameter values to search for. If search_range is an int/float, it is used as the global search range for parameters. This creates a search problem on the following space: \n",
    "\n",
    "optimization on (initial value / search_range, initial value * search_range)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "db4adade",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Values in \"params_dict\" are initial values (center of the search space) for the corresponding key\n",
    "params_dict = {'dropout': 0.3, 'learning_rate' : 0.001}\n",
    "\n",
    "# Radius of search space , must have the same Key as \"params_dict\"\n",
    "search_range = {'dropout': 2, 'learning_rate' : 5}   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ea5ba39f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluation \t Proposed point \t  Current eval. \t Best eval.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_14:0\", shape=(395,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_13:0\", shape=(395, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_17:0\", shape=(1154,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_16:0\", shape=(1154, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_20:0\", shape=(1161,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_19:0\", shape=(1161, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_23:0\", shape=(148,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_22:0\", shape=(148, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_11:0\", shape=(395,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_10:0\", shape=(395, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_13:0\", shape=(1154,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_12:0\", shape=(1154, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_15:0\", shape=(1161,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_14:0\", shape=(1161, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_17:0\", shape=(148,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_16:0\", shape=(148, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_14:0\", shape=(395,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_13:0\", shape=(395, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_17:0\", shape=(1154,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_16:0\", shape=(1154, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_20:0\", shape=(1161,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_19:0\", shape=(1161, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_23:0\", shape=(148,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_22:0\", shape=(148, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_14:0\", shape=(378,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_13:0\", shape=(378, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_17:0\", shape=(1114,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_16:0\", shape=(1114, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_20:0\", shape=(984,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_19:0\", shape=(984, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_23:0\", shape=(192,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_22:0\", shape=(192, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_11:0\", shape=(378,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_10:0\", shape=(378, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_13:0\", shape=(1114,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_12:0\", shape=(1114, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_15:0\", shape=(984,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_14:0\", shape=(984, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_17:0\", shape=(192,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_16:0\", shape=(192, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_14:0\", shape=(378,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_13:0\", shape=(378, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_17:0\", shape=(1114,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_16:0\", shape=(1114, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_20:0\", shape=(984,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_19:0\", shape=(984, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_23:0\", shape=(192,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_22:0\", shape=(192, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool_1/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_conv_1/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model/graph_pool/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_14:0\", shape=(352,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_13:0\", shape=(352, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_17:0\", shape=(1168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_16:0\", shape=(1168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_20:0\", shape=(1104,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_19:0\", shape=(1104, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_23:0\", shape=(112,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_22:0\", shape=(112, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_11:0\", shape=(352,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_10:0\", shape=(352, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_13:0\", shape=(1168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_12:0\", shape=(1168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_15:0\", shape=(1104,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_14:0\", shape=(1104, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_17:0\", shape=(112,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_16:0\", shape=(112, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_14:0\", shape=(352,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_13:0\", shape=(352, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_17:0\", shape=(1168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_16:0\", shape=(1168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_20:0\", shape=(1104,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_19:0\", shape=(1104, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_23:0\", shape=(112,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_22:0\", shape=(112, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_14:0\", shape=(369,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_13:0\", shape=(369, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_17:0\", shape=(1082,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_16:0\", shape=(1082, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_20:0\", shape=(987,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_19:0\", shape=(987, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_23:0\", shape=(176,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_22:0\", shape=(176, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_11:0\", shape=(369,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_10:0\", shape=(369, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_13:0\", shape=(1082,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_12:0\", shape=(1082, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_15:0\", shape=(987,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_14:0\", shape=(987, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_17:0\", shape=(176,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_16:0\", shape=(176, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_14:0\", shape=(369,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_13:0\", shape=(369, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_17:0\", shape=(1082,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_16:0\", shape=(1082, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_20:0\", shape=(987,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_19:0\", shape=(987, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_23:0\", shape=(176,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_22:0\", shape=(176, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_3/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_conv_3/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_1/graph_pool_2/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_14:0\", shape=(329,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_13:0\", shape=(329, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_17:0\", shape=(1184,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_16:0\", shape=(1184, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_20:0\", shape=(1023,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_19:0\", shape=(1023, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_23:0\", shape=(152,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_22:0\", shape=(152, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_11:0\", shape=(329,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_10:0\", shape=(329, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_13:0\", shape=(1184,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_12:0\", shape=(1184, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_15:0\", shape=(1023,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_14:0\", shape=(1023, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_17:0\", shape=(152,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_16:0\", shape=(152, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_14:0\", shape=(329,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_13:0\", shape=(329, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_17:0\", shape=(1184,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_16:0\", shape=(1184, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_20:0\", shape=(1023,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_19:0\", shape=(1023, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_23:0\", shape=(152,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_22:0\", shape=(152, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_14:0\", shape=(341,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_13:0\", shape=(341, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_17:0\", shape=(1224,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_16:0\", shape=(1224, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_20:0\", shape=(975,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_19:0\", shape=(975, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_23:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_22:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_11:0\", shape=(341,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_10:0\", shape=(341, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_13:0\", shape=(1224,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_12:0\", shape=(1224, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_15:0\", shape=(975,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_14:0\", shape=(975, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_17:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_16:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_14:0\", shape=(341,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_13:0\", shape=(341, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_17:0\", shape=(1224,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_16:0\", shape=(1224, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_20:0\", shape=(975,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_19:0\", shape=(975, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_23:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_22:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_5/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_conv_5/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_2/graph_pool_4/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:5 out of the last 5 calls to <function KerasModel._compute_model at 0x7fdb4fb1ba70> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
      "WARNING:tensorflow:6 out of the last 6 calls to <function KerasModel._compute_model at 0x7fdb4fb1ba70> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
      "init   \t [0.25759397 0.00164653]. \t  0.3365335144498499 \t 0.3365335144498499\n",
      "init   \t [5.81149627e-01 3.09823437e-04]. \t  -4.123225306815302 \t 0.3365335144498499\n",
      "init   \t [0.46691308 0.00346177]. \t  0.30977173323640483 \t 0.3365335144498499\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_14:0\", shape=(365,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_13:0\", shape=(365, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_17:0\", shape=(1294,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_16:0\", shape=(1294, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_20:0\", shape=(1269,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_19:0\", shape=(1269, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_23:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_22:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_11:0\", shape=(365,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_10:0\", shape=(365, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_13:0\", shape=(1294,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_12:0\", shape=(1294, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_15:0\", shape=(1269,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_14:0\", shape=(1269, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_17:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_16:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_14:0\", shape=(365,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_13:0\", shape=(365, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_17:0\", shape=(1294,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_16:0\", shape=(1294, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_20:0\", shape=(1269,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_19:0\", shape=(1269, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_23:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_22:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_14:0\", shape=(381,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_13:0\", shape=(381, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_17:0\", shape=(1076,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_16:0\", shape=(1076, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_20:0\", shape=(1071,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_19:0\", shape=(1071, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_23:0\", shape=(144,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_22:0\", shape=(144, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_11:0\", shape=(381,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_10:0\", shape=(381, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_13:0\", shape=(1076,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_12:0\", shape=(1076, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_15:0\", shape=(1071,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_14:0\", shape=(1071, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_17:0\", shape=(144,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_16:0\", shape=(144, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_14:0\", shape=(381,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_13:0\", shape=(381, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_17:0\", shape=(1076,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_16:0\", shape=(1076, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_20:0\", shape=(1071,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_19:0\", shape=(1071, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_23:0\", shape=(144,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_22:0\", shape=(144, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_7/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_conv_7/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_6/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1      \t [0.38088723 0.005     ]. \t  \u001b[92m0.5581399674502492\u001b[0m \t 0.5581399674502492\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_14:0\", shape=(359,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_13:0\", shape=(359, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_17:0\", shape=(1250,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_16:0\", shape=(1250, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_20:0\", shape=(1161,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_19:0\", shape=(1161, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_23:0\", shape=(156,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_22:0\", shape=(156, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_11:0\", shape=(359,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_10:0\", shape=(359, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_13:0\", shape=(1250,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_12:0\", shape=(1250, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_15:0\", shape=(1161,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_14:0\", shape=(1161, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_17:0\", shape=(156,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_16:0\", shape=(156, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_14:0\", shape=(359,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_13:0\", shape=(359, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_17:0\", shape=(1250,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_16:0\", shape=(1250, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_20:0\", shape=(1161,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_19:0\", shape=(1161, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_23:0\", shape=(156,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_22:0\", shape=(156, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_14:0\", shape=(334,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_13:0\", shape=(334, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_17:0\", shape=(1218,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_16:0\", shape=(1218, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_20:0\", shape=(1032,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_19:0\", shape=(1032, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_23:0\", shape=(124,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_22:0\", shape=(124, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_11:0\", shape=(334,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_10:0\", shape=(334, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_13:0\", shape=(1218,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_12:0\", shape=(1218, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_15:0\", shape=(1032,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_14:0\", shape=(1032, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_17:0\", shape=(124,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_16:0\", shape=(124, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_14:0\", shape=(334,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_13:0\", shape=(334, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_17:0\", shape=(1218,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_16:0\", shape=(1218, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_20:0\", shape=(1032,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_19:0\", shape=(1032, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_23:0\", shape=(124,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_22:0\", shape=(124, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_9/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_conv_9/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_4/graph_pool_8/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2      \t [0.42180757 0.005     ]. \t  0.42900747496414415 \t 0.5581399674502492\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_14:0\", shape=(328,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_13:0\", shape=(328, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_17:0\", shape=(1090,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_16:0\", shape=(1090, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_20:0\", shape=(1014,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_19:0\", shape=(1014, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_23:0\", shape=(100,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_22:0\", shape=(100, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_11:0\", shape=(328,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_10:0\", shape=(328, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_13:0\", shape=(1090,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_12:0\", shape=(1090, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_15:0\", shape=(1014,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_14:0\", shape=(1014, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_17:0\", shape=(100,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_16:0\", shape=(100, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_14:0\", shape=(328,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_13:0\", shape=(328, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_17:0\", shape=(1090,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_16:0\", shape=(1090, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_20:0\", shape=(1014,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_19:0\", shape=(1014, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_23:0\", shape=(100,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_22:0\", shape=(100, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_14:0\", shape=(352,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_13:0\", shape=(352, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_17:0\", shape=(1136,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_16:0\", shape=(1136, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_20:0\", shape=(990,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_19:0\", shape=(990, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_23:0\", shape=(144,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_22:0\", shape=(144, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_11:0\", shape=(352,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_10:0\", shape=(352, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_13:0\", shape=(1136,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_12:0\", shape=(1136, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_15:0\", shape=(990,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_14:0\", shape=(990, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_17:0\", shape=(144,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_16:0\", shape=(144, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_14:0\", shape=(352,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_13:0\", shape=(352, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_17:0\", shape=(1136,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_16:0\", shape=(1136, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_20:0\", shape=(990,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_19:0\", shape=(990, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_23:0\", shape=(144,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_22:0\", shape=(144, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_11/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_conv_11/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_5/graph_pool_10/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3      \t [3.5306479e-01 2.0000000e-04]. \t  -1.2687765174239356 \t 0.5581399674502492\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_14:0\", shape=(365,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_13:0\", shape=(365, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_17:0\", shape=(1194,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_16:0\", shape=(1194, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_20:0\", shape=(999,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_19:0\", shape=(999, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_23:0\", shape=(168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_22:0\", shape=(168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_11:0\", shape=(365,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_10:0\", shape=(365, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_13:0\", shape=(1194,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_12:0\", shape=(1194, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_15:0\", shape=(999,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_14:0\", shape=(999, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_17:0\", shape=(168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_16:0\", shape=(168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_14:0\", shape=(365,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_13:0\", shape=(365, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_17:0\", shape=(1194,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_16:0\", shape=(1194, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_20:0\", shape=(999,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_19:0\", shape=(999, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_23:0\", shape=(168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_22:0\", shape=(168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_14:0\", shape=(340,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_13:0\", shape=(340, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_17:0\", shape=(1200,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_16:0\", shape=(1200, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_20:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_19:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_23:0\", shape=(116,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_22:0\", shape=(116, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_11:0\", shape=(340,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_10:0\", shape=(340, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_13:0\", shape=(1200,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_12:0\", shape=(1200, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_15:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_14:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_17:0\", shape=(116,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_16:0\", shape=(116, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_14:0\", shape=(340,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_13:0\", shape=(340, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_17:0\", shape=(1200,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_16:0\", shape=(1200, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_20:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_19:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_23:0\", shape=(116,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_22:0\", shape=(116, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_13/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_conv_13/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_6/graph_pool_12/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4      \t [0.15  0.005]. \t  \u001b[92m0.6009713994268596\u001b[0m \t 0.6009713994268596\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_14:0\", shape=(361,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_13:0\", shape=(361, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_17:0\", shape=(1176,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_16:0\", shape=(1176, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_20:0\", shape=(1197,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_19:0\", shape=(1197, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_23:0\", shape=(136,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_22:0\", shape=(136, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_11:0\", shape=(361,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_10:0\", shape=(361, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_13:0\", shape=(1176,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_12:0\", shape=(1176, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_15:0\", shape=(1197,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_14:0\", shape=(1197, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_17:0\", shape=(136,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_16:0\", shape=(136, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_14:0\", shape=(361,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_13:0\", shape=(361, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_17:0\", shape=(1176,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_16:0\", shape=(1176, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_20:0\", shape=(1197,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_19:0\", shape=(1197, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_23:0\", shape=(136,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_22:0\", shape=(136, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_14:0\", shape=(379,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_13:0\", shape=(379, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_17:0\", shape=(1180,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_16:0\", shape=(1180, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_20:0\", shape=(1125,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_19:0\", shape=(1125, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_11:0\", shape=(379,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_10:0\", shape=(379, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_13:0\", shape=(1180,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_12:0\", shape=(1180, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_15:0\", shape=(1125,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_14:0\", shape=(1125, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_17:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_16:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_14:0\", shape=(379,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_13:0\", shape=(379, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_17:0\", shape=(1180,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_16:0\", shape=(1180, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_20:0\", shape=(1125,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_19:0\", shape=(1125, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_15/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_conv_15/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_7/graph_pool_14/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5      \t [0.2144198 0.005    ]. \t  0.549538962337399 \t 0.6009713994268596\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_14:0\", shape=(323,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_13:0\", shape=(323, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_17:0\", shape=(1162,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_16:0\", shape=(1162, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_20:0\", shape=(1011,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_19:0\", shape=(1011, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_23:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_22:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_11:0\", shape=(323,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_10:0\", shape=(323, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_13:0\", shape=(1162,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_12:0\", shape=(1162, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_15:0\", shape=(1011,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_14:0\", shape=(1011, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_17:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_16:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_14:0\", shape=(323,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_13:0\", shape=(323, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_17:0\", shape=(1162,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_16:0\", shape=(1162, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_20:0\", shape=(1011,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_19:0\", shape=(1011, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_23:0\", shape=(128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_22:0\", shape=(128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_14:0\", shape=(340,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_13:0\", shape=(340, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_17:0\", shape=(1060,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_16:0\", shape=(1060, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_20:0\", shape=(990,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_19:0\", shape=(990, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_23:0\", shape=(76,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_22:0\", shape=(76, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_11:0\", shape=(340,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_10:0\", shape=(340, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_13:0\", shape=(1060,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_12:0\", shape=(1060, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_15:0\", shape=(990,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_14:0\", shape=(990, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_17:0\", shape=(76,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_16:0\", shape=(76, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_14:0\", shape=(340,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_13:0\", shape=(340, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_17:0\", shape=(1060,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_16:0\", shape=(1060, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_20:0\", shape=(990,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_19:0\", shape=(990, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_23:0\", shape=(76,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_22:0\", shape=(76, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_17/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_conv_17/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_8/graph_pool_16/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6      \t [0.39398148 0.005     ]. \t  0.4722427643242765 \t 0.6009713994268596\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_14:0\", shape=(363,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_13:0\", shape=(363, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_17:0\", shape=(1248,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_16:0\", shape=(1248, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_20:0\", shape=(1149,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_19:0\", shape=(1149, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_23:0\", shape=(168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_22:0\", shape=(168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_11:0\", shape=(363,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_10:0\", shape=(363, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_13:0\", shape=(1248,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_12:0\", shape=(1248, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_15:0\", shape=(1149,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_14:0\", shape=(1149, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_17:0\", shape=(168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_16:0\", shape=(168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_14:0\", shape=(363,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_13:0\", shape=(363, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_17:0\", shape=(1248,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_16:0\", shape=(1248, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_20:0\", shape=(1149,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_19:0\", shape=(1149, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_23:0\", shape=(168,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_22:0\", shape=(168, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_14:0\", shape=(355,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_13:0\", shape=(355, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_17:0\", shape=(1218,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_16:0\", shape=(1218, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_20:0\", shape=(1023,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_19:0\", shape=(1023, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_23:0\", shape=(180,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_22:0\", shape=(180, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_11:0\", shape=(355,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_10:0\", shape=(355, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_13:0\", shape=(1218,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_12:0\", shape=(1218, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_15:0\", shape=(1023,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_14:0\", shape=(1023, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_17:0\", shape=(180,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_16:0\", shape=(180, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_14:0\", shape=(355,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_13:0\", shape=(355, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_17:0\", shape=(1218,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_16:0\", shape=(1218, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_20:0\", shape=(1023,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_19:0\", shape=(1023, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_23:0\", shape=(180,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_22:0\", shape=(180, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_19/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_conv_19/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_9/graph_pool_18/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7      \t [0.24706651 0.005     ]. \t  0.18921841526205152 \t 0.6009713994268596\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_14:0\", shape=(334,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_13:0\", shape=(334, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_17:0\", shape=(1244,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_16:0\", shape=(1244, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_20:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_19:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_23:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_22:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_11:0\", shape=(334,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_10:0\", shape=(334, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_13:0\", shape=(1244,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_12:0\", shape=(1244, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_15:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_14:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_17:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_16:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_14:0\", shape=(334,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_13:0\", shape=(334, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_17:0\", shape=(1244,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_16:0\", shape=(1244, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_20:0\", shape=(1098,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_19:0\", shape=(1098, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_23:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_22:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_14:0\", shape=(314,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_13:0\", shape=(314, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_17:0\", shape=(1184,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_16:0\", shape=(1184, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_20:0\", shape=(984,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_19:0\", shape=(984, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_23:0\", shape=(108,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_22:0\", shape=(108, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_11:0\", shape=(314,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_10:0\", shape=(314, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_13:0\", shape=(1184,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_12:0\", shape=(1184, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_15:0\", shape=(984,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_14:0\", shape=(984, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_17:0\", shape=(108,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_16:0\", shape=(108, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_14:0\", shape=(314,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_13:0\", shape=(314, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_17:0\", shape=(1184,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_16:0\", shape=(1184, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_20:0\", shape=(984,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_19:0\", shape=(984, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_23:0\", shape=(108,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_22:0\", shape=(108, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_21/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_conv_21/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_10/graph_pool_20/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8      \t [0.1847394 0.0002   ]. \t  0.49346601531255485 \t 0.6009713994268596\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_14:0\", shape=(300,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_13:0\", shape=(300, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_17:0\", shape=(1290,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_16:0\", shape=(1290, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_20:0\", shape=(1014,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_19:0\", shape=(1014, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_23:0\", shape=(116,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_22:0\", shape=(116, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_11:0\", shape=(300,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_10:0\", shape=(300, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_13:0\", shape=(1290,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_12:0\", shape=(1290, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_15:0\", shape=(1014,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_14:0\", shape=(1014, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_17:0\", shape=(116,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_16:0\", shape=(116, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_14:0\", shape=(300,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_13:0\", shape=(300, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_17:0\", shape=(1290,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_16:0\", shape=(1290, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_20:0\", shape=(1014,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_19:0\", shape=(1014, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_23:0\", shape=(116,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_22:0\", shape=(116, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_14:0\", shape=(383,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_13:0\", shape=(383, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_17:0\", shape=(1200,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_16:0\", shape=(1200, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_20:0\", shape=(1077,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_19:0\", shape=(1077, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_23:0\", shape=(196,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_22:0\", shape=(196, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_11:0\", shape=(383,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_10:0\", shape=(383, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_13:0\", shape=(1200,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_12:0\", shape=(1200, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_15:0\", shape=(1077,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_14:0\", shape=(1077, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_17:0\", shape=(196,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_16:0\", shape=(196, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_14:0\", shape=(383,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_13:0\", shape=(383, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_17:0\", shape=(1200,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_16:0\", shape=(1200, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_20:0\", shape=(1077,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_19:0\", shape=(1077, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_23:0\", shape=(196,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_22:0\", shape=(196, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_23/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_conv_23/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_11/graph_pool_22/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9      \t [0.15387001 0.00366433]. \t  0.5786634529026986 \t 0.6009713994268596\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_14:0\", shape=(368,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_13:0\", shape=(368, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_17:0\", shape=(980,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_16:0\", shape=(980, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_20:0\", shape=(942,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_19:0\", shape=(942, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_11:0\", shape=(368,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_10:0\", shape=(368, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_13:0\", shape=(980,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_12:0\", shape=(980, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_15:0\", shape=(942,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_14:0\", shape=(942, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_17:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_16:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_19:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_18:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_21:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_20:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_23:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_22:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_25:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_24:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_27:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_26:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_8:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_29:0\", shape=(0,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_28:0\", shape=(0, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_9:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_14:0\", shape=(368,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_13:0\", shape=(368, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_17:0\", shape=(980,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_16:0\", shape=(980, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_20:0\", shape=(942,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_19:0\", shape=(942, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_23:0\", shape=(160,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_22:0\", shape=(160, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_14:0\", shape=(337,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_13:0\", shape=(337, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_17:0\", shape=(1156,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_16:0\", shape=(1156, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_20:0\", shape=(1119,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_19:0\", shape=(1119, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_23:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_22:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_11:0\", shape=(337,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_10:0\", shape=(337, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_13:0\", shape=(1156,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_12:0\", shape=(1156, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_15:0\", shape=(1119,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_14:0\", shape=(1119, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_17:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_16:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_14:0\", shape=(337,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_13:0\", shape=(337, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_17:0\", shape=(1156,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_16:0\", shape=(1156, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_20:0\", shape=(1119,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_19:0\", shape=(1119, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_23:0\", shape=(120,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_22:0\", shape=(120, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_25/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_11:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_10:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_13:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_12:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_1:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_15:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_14:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_2:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_conv_25/Cast_3:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_12/graph_pool_24/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10     \t [0.20384127 0.0038913 ]. \t  0.5100267062642383 \t 0.6009713994268596\n"
     ]
    }
   ],
   "source": [
    "# We don't want to optimize all parameters. Some parameters like 'mode' are fixed and defined below:\n",
    "optimizer = dc.hyper.GaussianProcessHyperparamOpt(\n",
    "    lambda **p: dc.models.GraphConvModel(n_tasks=n_tasks, mode=\"regression\", **p))\n",
    "\n",
    "best_model, best_params, all_results = optimizer.hyperparam_search(\n",
    "            params_dict,       # Dictionary that maps hyperparameter names (strings) to the possible parameter values \n",
    "            train_dataset, \n",
    "            valid_dataset,     # Dataset used for validation (optimization on valid scores \n",
    "            metric_r2,         # Metric used for evaluation\n",
    "            nb_epoch=50,\n",
    "            use_max=True,      # If True, return the model with the highest score. Else return model with the minimum score.\n",
    "            logdir=save_hyperparms,   # Directory in which to store created models. If not set, will use a temporary directory.    \n",
    "            max_iter=10,        # No. of optimization trials\n",
    "            search_range=search_range\n",
    "            )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0151c3bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'dropout': 0.15, 'learning_rate': 0.005}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "111fe5cc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'_dropout_0.257594_learning_rate_0.001647': 0.3365335144498499,\n",
       " '_dropout_0.581150_learning_rate_0.000310': -4.123225306815302,\n",
       " '_dropout_0.466913_learning_rate_0.003462': 0.30977173323640483,\n",
       " '_dropout_0.380887_learning_rate_0.005000': 0.5581399674502492,\n",
       " '_dropout_0.421808_learning_rate_0.005000': 0.42900747496414415,\n",
       " '_dropout_0.353065_learning_rate_0.000200': -1.2687765174239356,\n",
       " '_dropout_0.150000_learning_rate_0.005000': 0.6009713994268596,\n",
       " '_dropout_0.214420_learning_rate_0.005000': 0.549538962337399,\n",
       " '_dropout_0.393981_learning_rate_0.005000': 0.4722427643242765,\n",
       " '_dropout_0.247067_learning_rate_0.005000': 0.18921841526205152,\n",
       " '_dropout_0.184739_learning_rate_0.000200': 0.49346601531255485,\n",
       " '_dropout_0.153870_learning_rate_0.003664': 0.5786634529026986,\n",
       " '_dropout_0.203841_learning_rate_0.003891': 0.5100267062642383}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "75654392",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['./result/Hyperparameter/model_dropout_0.150000_learning_rate_0.005000/ckpt-1']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model.get_checkpoints()   # filename of best model saved in the folder \"save_hyperparms\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76109ed2",
   "metadata": {},
   "source": [
    "### Load saved best model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "04ead59e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'./result/Hyperparameter/model_dropout_0.150000_learning_rate_0.005000'"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model_checkpoints = best_model.get_checkpoints()[0]\n",
    "best_model_folder = os.path.dirname(best_model_checkpoints)   # name of folder where best model is saved\n",
    "best_model_folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a1de5a50",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize model with best hyperparameters\n",
    "model = dc.models.GraphConvModel(n_tasks=1, mode='regression', \n",
    "                  dropout=best_params['dropout'], learning_rate=best_params['learning_rate'],\n",
    "                  model_dir=best_model_folder)\n",
    "\n",
    "model.restore()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "e261d53c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: {'r2_score': 0.7352561440217054}\n",
      "Validation set score: {'r2_score': 0.6009713994268596}\n",
      "Test set score: {'r2_score': 0.6062390140581315}\n"
     ]
    }
   ],
   "source": [
    "print(\"Training set score:\", model.evaluate(train_dataset, metric_r2))\n",
    "print(\"Validation set score:\", model.evaluate(valid_dataset, metric_r2))\n",
    "print(\"Test set score:\", model.evaluate(test_dataset, metric_r2))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e52b47e3",
   "metadata": {},
   "source": [
    "# Early Stopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "22f0e90a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Learning Rate Schedules (we don't use a fixed value for learning rate\n",
    "# The learning rate starts as initial_rate. Every decay_steps training steps, it is multiplied by decay_rate.\n",
    "# https://deepchem.readthedocs.io/en/latest/api_reference/models.html#deepchem.models.optimizers.ExponentialDecay\n",
    "initial_rate = 0.001\n",
    "decay_rate = 0.9\n",
    "decay_steps = 200\n",
    "learning_rate = dc.models.optimizers.ExponentialDecay(initial_rate, decay_rate, decay_steps)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32328add",
   "metadata": {},
   "source": [
    "#### We set \"tensorboard=True\" to log metric of validation dataset during training. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "e2ca2ee8",
   "metadata": {},
   "outputs": [],
   "source": [
    "del model\n",
    "\n",
    "model = dc.models.GraphConvModel(n_tasks=n_tasks,              # No. of tasks \n",
    "                                 graph_conv_layers=[64],       # Width of channels for the Graph Convolution Layers\n",
    "                                 dense_layer_size=128,         # Width of channels for Atom Level Dense Layer before GraphPool\n",
    "                                 dropout=0.2,                  # Dropout probablity to use for each layer. The length of this list should equal len(graph_conv_layers)+1 (one value for each convolution layer, and one for the dense layer). Alternatively this may be a single value instead of a list, in which case the same value is used for every layer.\n",
    "                                 mode='regression',            # Either “classification” or “regression”\n",
    "                                 batch_size=100,               # Batch size for training and evaluating\n",
    "                                 model_dir=model_dir,          # Directory on disk where the model will be stored. If this is None, a temporary directory is created.\n",
    "                                 learning_rate=learning_rate,\n",
    "                                 tensorboard=True              # log progress to TensorBoard during training                                 \n",
    "                                 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "704d7b53",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define \"callback\" function for early stopping\n",
    "# It computes the validation set's metric every \"interval\" training steps (not epoch). \n",
    "interval = 100\n",
    "callback = dc.models.ValidationCallback(valid_dataset, interval, metric_r2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "f4b0094f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_14:0\", shape=(354,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_13:0\", shape=(354, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_17:0\", shape=(1214,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_16:0\", shape=(1214, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_20:0\", shape=(1128,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_19:0\", shape=(1128, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_23:0\", shape=(176,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_22:0\", shape=(176, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_14:0\", shape=(349,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_13:0\", shape=(349, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_17:0\", shape=(1204,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_16:0\", shape=(1204, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_20:0\", shape=(1023,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_19:0\", shape=(1023, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_23:0\", shape=(200,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_22:0\", shape=(200, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_14:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_13:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_4:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_17:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_16:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_5:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_20:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_19:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_6:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n",
      "/opt/anaconda3/envs/GNN/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:449: UserWarning: Converting sparse IndexedSlices(IndexedSlices(indices=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_23:0\", shape=(None,), dtype=int32), values=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Reshape_22:0\", shape=(None, 64), dtype=float32), dense_shape=Tensor(\"gradient_tape/private__graph_conv_keras_model_3/graph_pool_3/Cast_7:0\", shape=(2,), dtype=int32))) to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"shape. This may consume a large amount of memory.\" % value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Step 100 validation: r2_score=0.413665\n",
      "Step 200 validation: r2_score=0.398329\n",
      "Step 300 validation: r2_score=0.514176\n",
      "Step 400 validation: r2_score=0.541485\n",
      "Step 500 validation: r2_score=0.602379\n",
      "Step 600 validation: r2_score=0.634528\n",
      "Step 700 validation: r2_score=0.609275\n",
      "Step 800 validation: r2_score=0.61514\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.4439974212646484"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(train_dataset, nb_epoch=100, callbacks=callback)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "02388c45",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training set score: {'r2_score': 0.7951934295059822}\n",
      "validation set score: {'r2_score': 0.615140117595991}\n",
      "test set score: {'r2_score': 0.6539251579008657}\n"
     ]
    }
   ],
   "source": [
    "print('training set score:', model.evaluate(train_dataset, metric_r2))\n",
    "print('validation set score:', model.evaluate(valid_dataset, metric_r2))\n",
    "print('test set score:', model.evaluate(test_dataset, metric_r2))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95378710",
   "metadata": {},
   "source": [
    "### Tensorboard\n",
    "\n",
    "In order to see tensorboard, open Terminal. Go to the folder where you have saved this notebook by \"cd\" command. Then run the following command:\n",
    "\n",
    "tensorboard --logdir=result\n",
    "\n",
    "In the above command, \"result\" is the name of folder that model is saved.It is the value of parameter \"model_dir\". We set parameter \"model_dir\" when we initialized the model. \n",
    "Remember not to use '' marks in the above command for folder name. \n",
    "\n",
    "If you change the name of folder (set in the parameter \"model_dir\"), update the name of folder in the above command. \n",
    "\n",
    "This will launch the tensorboard web server on your local computer on port 6006. Open a new tab in your browser, and go to http://localhost:6006 in your web browser to look through tensorboard's UI."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e63a1ed8",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}