heavy_tail_bert_mrpc.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check clipped-SSTM and clipped-SGD on BERT + MRPCtask"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from graph_plot_tools import *\n",
    "from utils import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of examples:  5696\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kesha/.local/lib/python3.8/site-packages/pandas/util/_decorators.py:311: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n",
      "  return func(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>#1 ID</th>\n",
       "      <th>#2 ID</th>\n",
       "      <th>sent1</th>\n",
       "      <th>sent2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2750203</td>\n",
       "      <td>2750153</td>\n",
       "      <td>Officials at Brandeis said this was an \" extre...</td>\n",
       "      <td>\" This is an extremely heartrending time for t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2622625</td>\n",
       "      <td>2622698</td>\n",
       "      <td>\" If that ain 't a Democrat , I must be at the...</td>\n",
       "      <td>And if that ain 't a Democrat , then I must be...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1946553</td>\n",
       "      <td>1946598</td>\n",
       "      <td>Fewer than a dozen FBI agents were dispatched ...</td>\n",
       "      <td>Fewer than a dozen FBI agents will be sent to ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>2216702</td>\n",
       "      <td>2216672</td>\n",
       "      <td>Those who only had surgery lived an average of...</td>\n",
       "      <td>For those who got surgery alone , median survi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>742090</td>\n",
       "      <td>742015</td>\n",
       "      <td>Tonight a spokesman for Russia 's foreign mini...</td>\n",
       "      <td>Tonight a spokesman for the Russian Foreign Mi...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   label    #1 ID    #2 ID                                              sent1  \\\n",
       "0      1  2750203  2750153  Officials at Brandeis said this was an \" extre...   \n",
       "1      0  2622625  2622698  \" If that ain 't a Democrat , I must be at the...   \n",
       "2      1  1946553  1946598  Fewer than a dozen FBI agents were dispatched ...   \n",
       "3      1  2216702  2216672  Those who only had surgery lived an average of...   \n",
       "4      1   742090   742015  Tonight a spokesman for Russia 's foreign mini...   \n",
       "\n",
       "                                               sent2  \n",
       "0  \" This is an extremely heartrending time for t...  \n",
       "1  And if that ain 't a Democrat , then I must be...  \n",
       "2  Fewer than a dozen FBI agents will be sent to ...  \n",
       "3  For those who got surgery alone , median survi...  \n",
       "4  Tonight a spokesman for the Russian Foreign Mi...  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "data_folder_name = \"glue_data\"\n",
    "df1 = pd.read_table(\"./{}/MRPC/msr_paraphrase_train.txt\".format(data_folder_name), delimiter='\\\\t', )\n",
    "df2 = pd.read_table(\"./{}/MRPC/msr_paraphrase_test.txt\".format(data_folder_name), delimiter='\\\\t', )\n",
    "df = pd.concat([df1, df2]).rename(columns={'Quality': 'label', '#1 String': 'sent1', '#2 String': 'sent2'})\n",
    "print(\"number of examples: \", df.shape[0])\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading BERT tokenizer...\n"
     ]
    }
   ],
   "source": [
    "from transformers import BertTokenizer\n",
    "# Load the BERT tokenizer.\n",
    "print('Loading BERT tokenizer...')\n",
    "tokenizer = BertTokenizer.from_pretrained('bert-base-uncased', do_lower_case=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original 1:  Officials at Brandeis said this was an \" extremely heartrending \" time for the campus .\n",
      "Original 2:  \" This is an extremely heartrending time for the entire Brandeis University community .\n",
      "Token IDs: [101, 4584, 2012, 4435, 17580, 2056, 2023, 2001, 2019, 1000, 5186, 2540, 7389, 4667, 1000, 2051, 2005, 1996, 3721, 1012, 102, 1000, 2023, 2003, 2019, 5186, 2540, 7389, 4667, 2051, 2005, 1996, 2972, 4435, 17580, 2118, 2451, 1012, 102]\n"
     ]
    }
   ],
   "source": [
    "# Tokenize all of the sentences and map the tokens to their word IDs.\n",
    "input_ids = []\n",
    "# For every sentence...\n",
    "for (sent1, sent2) in zip(df[\"sent1\"].values, df[\"sent2\"].values):\n",
    "    sent = (sent1, sent2)\n",
    "    # `encode` will:\n",
    "    #   (1) Tokenize the sentence.\n",
    "    #   (2) Prepend the `[CLS]` token to the start.\n",
    "    #   (3) Append the `[SEP]` token to the end.\n",
    "    #   (4) Map tokens to their IDs.\n",
    "    encoded_sent = tokenizer.encode(\n",
    "                        *sent,                      # Sentence to encode.\n",
    "                        add_special_tokens = True, # Add '[CLS]' and '[SEP]'\n",
    "                        # This function also supports truncation and conversion\n",
    "                        # to pytorch tensors, but we need to do padding, so we\n",
    "                        # can't use these features :( .\n",
    "                        #max_length = 128,          # Truncate all sentences.\n",
    "                        #return_tensors = 'pt',     # Return pytorch tensors.\n",
    "                   )\n",
    "    \n",
    "    # Add the encoded sentence to the list.\n",
    "    input_ids.append(encoded_sent)\n",
    "# Print sentence 0, now as a list of IDs.\n",
    "print('Original 1: ', df[\"sent1\"].values[0])\n",
    "print('Original 2: ', df[\"sent2\"].values[0])\n",
    "print('Token IDs:', input_ids[0])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Max sentence length:  104\n"
     ]
    }
   ],
   "source": [
    "print('Max sentence length: ', max([len(sen) for sen in input_ids]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Padding/truncating all sentences to 105 values...\n",
      "\n",
      "Padding token: \"[PAD]\", ID: 0\n",
      "\\Done.\n"
     ]
    }
   ],
   "source": [
    "# from keras.preprocessing.sequence import pad_sequences\n",
    "def pad_sequences(seqs, maxlen=None, value=0, padding=\"post\"):\n",
    "    if maxlen is None:\n",
    "        raise ValueError(\"Invalid maxlen: {}\".format(maxlen))\n",
    "    for i in range(len(seqs)):\n",
    "        add = [value] * max(0, maxlen - len(seqs[i]))\n",
    "        if padding == \"post\":\n",
    "            seqs[i] = seqs[i] + add\n",
    "        elif padding == \"pre\":\n",
    "            seqs[i] = add + seqs[i]\n",
    "    return seqs\n",
    "\n",
    "MAX_LEN =  max([len(sen) for sen in input_ids])+1\n",
    "print('\\nPadding/truncating all sentences to %d values...' % MAX_LEN)\n",
    "print('\\nPadding token: \"{:}\", ID: {:}'.format(tokenizer.pad_token, tokenizer.pad_token_id))\n",
    "\n",
    "input_ids = pad_sequences(input_ids, maxlen=MAX_LEN, value=0, padding=\"post\")\n",
    "print('\\Done.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create attention masks\n",
    "attention_masks = []\n",
    "# For each sentence...\n",
    "for sent in input_ids:\n",
    "    \n",
    "    # Create the attention mask.\n",
    "    #   - If a token ID is 0, then it's padding, set the mask to 0.\n",
    "    #   - If a token ID is > 0, then it's a real token, set the mask to 1.\n",
    "    att_mask = [int(token_id > 0) for token_id in sent]\n",
    "    \n",
    "    # Store the attention mask for this sentence.\n",
    "    attention_masks.append(att_mask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x7f3c3b8ef760>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.pcolor(attention_masks[0:2000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 1, ..., 0, 1, 1])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.label.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset train/val splits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import matthews_corrcoef\n",
    "\n",
    "outer_constant = 21\n",
    "\n",
    "data_split_rs = outer_constant\n",
    "\n",
    "train_inputs, validation_inputs, train_labels, validation_labels = train_test_split(\n",
    "    input_ids,\n",
    "    df.label.values, \n",
    "    random_state=data_split_rs, \n",
    "    test_size=0.1\n",
    ")\n",
    "\n",
    "train_masks, validation_masks, _, _ = train_test_split(\n",
    "    attention_masks, \n",
    "    df.label.values, \n",
    "    random_state=data_split_rs, \n",
    "    test_size=0.1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 1 GPU(s) available.\n",
      "We will use the GPU: NVIDIA GeForce RTX 2070 SUPER\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import scipy.stats\n",
    "import matplotlib.pyplot as plt\n",
    "import tqdm\n",
    "import numpy as np\n",
    "\n",
    "# If there's a GPU available...\n",
    "if torch.cuda.is_available():    \n",
    "    # Tell PyTorch to use the GPU.    \n",
    "    device = torch.device(\"cuda\")\n",
    "    print('There are %d GPU(s) available.' % torch.cuda.device_count())\n",
    "    print('We will use the GPU:', torch.cuda.get_device_name(0))\n",
    "# If not...\n",
    "else:\n",
    "    print('No GPU available, using the CPU instead.')\n",
    "    device = torch.device(\"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_inputs = torch.tensor(train_inputs)\n",
    "validation_inputs = torch.tensor(validation_inputs)\n",
    "train_labels = torch.tensor(train_labels)\n",
    "validation_labels = torch.tensor(validation_labels)\n",
    "train_masks = torch.tensor(train_masks)\n",
    "validation_masks = torch.tensor(validation_masks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_seed = 1\n",
    "num_workers = 1\n",
    "\n",
    "from torch.utils.data import TensorDataset, DataLoader, RandomSampler, SequentialSampler\n",
    "###full_grad_dataloader\n",
    "batch_size = 32 # train_inputs.shape[0]\n",
    "\n",
    "# Create the DataLoader for our training set.\n",
    "train_data = TensorDataset(train_inputs, train_masks, train_labels)\n",
    "train_sampler = RandomSampler(train_data)\n",
    "train_loader_for_full_grad = DataLoader(\n",
    "    train_data, batch_size=batch_size, num_workers=num_workers,\n",
    "    worker_init_fn = lambda id: np.random.seed(id + num_workers * random_seed)\n",
    ")\n",
    "\n",
    "\n",
    "### for training\n",
    "batch_size = 32 # train_inputs.shape[0]\n",
    "# Create the DataLoader for our training set.\n",
    "train_data = TensorDataset(train_inputs, train_masks, train_labels)\n",
    "train_loader = DataLoader(\n",
    "    train_data, batch_size=batch_size, shuffle=True, num_workers=num_workers,\n",
    "    worker_init_fn = lambda id: np.random.seed(id + num_workers * random_seed)\n",
    ")\n",
    "\n",
    "validation_data = TensorDataset(validation_inputs, validation_masks, validation_labels)\n",
    "val_loader = DataLoader(\n",
    "    validation_data, batch_size=batch_size, shuffle=True, num_workers=num_workers,\n",
    "    worker_init_fn = lambda id: np.random.seed(id + num_workers * random_seed)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tasks (models + optimizers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# from transformers import BertForSequenceClassification\n",
    "# BertForSequenceClassification.from_pretrained(\n",
    "#     \"bert-base-uncased\", # Use the 12-layer BERT model, with an uncased vocab.\n",
    "#     num_labels = 2, # The number of output labels--2 for binary classification.\n",
    "#                     # You can increase this for multi-class tasks.   \n",
    "#     output_attentions = False, # Whether the model returns attentions weights.\n",
    "#     output_hidden_states = False, # Whether the model returns all hidden-states.\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# of setups: 5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.bias', 'cls.predictions.decoder.weight']\n",
      "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    }
   ],
   "source": [
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "\n",
    "from transformers import BertForSequenceClassification\n",
    "\n",
    "from optimizers import clipped_SGD, clipped_SSTM\n",
    "\n",
    "nets = []\n",
    "\n",
    "# lrs = [1e-3, 3e-3, 1e-2]\n",
    "# bs_muls_list = [2, 4, 8]\n",
    "# batch_mul_step_count_list = [50, 100, 200]\n",
    "\n",
    "history_random_seed = outer_constant\n",
    "# shift_mul = len(lrs) * len(batch_mul_step_count_list)\n",
    "shift_mul = 5\n",
    "\n",
    "print(\"# of setups: {}\".format(shift_mul))\n",
    "\n",
    "starting_point_random_seed_shift = outer_constant * shift_mul\n",
    "\n",
    "for starting_point_random_seed in range(5):\n",
    "    for _ in range(starting_point_random_seed_shift, starting_point_random_seed_shift + shift_mul):\n",
    "            # setting the same starting point for all algorithms\n",
    "            torch.manual_seed(starting_point_random_seed)\n",
    "            nets.append(\n",
    "                BertForSequenceClassification.from_pretrained(\n",
    "                    \"bert-base-uncased\", # Use the 12-layer BERT model, with an uncased vocab.\n",
    "                    num_labels = 2, # The number of output labels--2 for binary classification.\n",
    "                                    # You can increase this for multi-class tasks.   \n",
    "                    output_attentions = False, # Whether the model returns attentions weights.\n",
    "                    output_hidden_states = False, # Whether the model returns all hidden-states.\n",
    "                )\n",
    "            )\n",
    "\n",
    "            nets[-1].zero_grad()\n",
    "            for i, name_param in enumerate(nets[-1].named_parameters()):\n",
    "                if i < 197: # 197 -- 2 linears, 199 - only last linear\n",
    "                    name_param[1].requires_grad = False\n",
    "\n",
    "            nets[-1].train()\n",
    "\n",
    "# changing this one doesn't affect starting point, but changes training process (history)\n",
    "torch.manual_seed(history_random_seed)\n",
    "\n",
    "\n",
    "opts = []\n",
    "\n",
    "bs_muls = []\n",
    "\n",
    "batch_mul_step_counts = []\n",
    "\n",
    "import itertools\n",
    "\n",
    "for i in range(5):\n",
    "#     for j, (lr, batch_mul_step_count) in enumerate(itertools.product(lrs, batch_mul_step_count_list)):\n",
    "    bs_muls += [1, 1, 1, 4, \"linear\"]\n",
    "    batch_mul_step_counts += [400, 400, 400, 400, 100]\n",
    "    opts += [\n",
    "        torch.optim.Adam(nets[i * 5 + 0].parameters(), lr = 5e-4),\n",
    "        optim.SGD(nets[i * 5 + 1].parameters(), lr=3e-4, momentum=0.99),\n",
    "        clipped_SGD(nets[i * 5 + 2].parameters(), lr=1e-3, momentum=0.95,\n",
    "                       clipping_type=\"coordinate_wise\", clipping_level=0.1),\n",
    "        clipped_SSTM(nets[i * 5 + 3].parameters(), lr=3e-3, L=1, \n",
    "                     clipping_type='norm', clipping_level=1.0, nu=1),  # with bs_mul = 4\n",
    "        clipped_SSTM(nets[i * 5 + 4].parameters(), lr=1e-2, L=1, \n",
    "                    clipping_type='norm', clipping_level=1.0, nu=1),  # with bs_mul = linear (100)\n",
    "    ]\n",
    "    \n",
    "# [\n",
    "#     torch.optim.Adam(nets[i * 5 + 3].parameters(), lr = 5e-4),\n",
    "#     \"Adam, 5e-4, bs_mul = 1\",\n",
    "#     optim.SGD(nets[i * 3 + 0].parameters(), lr=1e-3, momentum=0.9),\n",
    "#     \"SGD, 1e-3, 0.9, bs_mul = 1\",\n",
    "#     clipped_SGD(nets[i * 5 + 3].parameters(), lr=2e-3, momentum=0.9, \n",
    "#                 clipping_type=\"coordinate_wise\", clipping_level=0.1),\n",
    "#     \"clipped_SGD, 2e-3, 0.9, coordinate_wise, 0.1, bs_mul = 1\",\n",
    "#     clipped_SSTM(nets[i * 5 + 2].parameters(), lr=1e-3, L=1, \n",
    "#                  clipping_type='norm', clipping_level=1.0, nu=1),  # with bs_mul = 8\n",
    "#     \"clipped_SSTM, 1e-3, 1, norm, 1, bs_mul = 8, nu=1\",\n",
    "#     clipped_SSTM(nets[i * 5 + 2].parameters(), lr=1e-3, L=1, \n",
    "#                 clipping_type='norm', clipping_level=1.0, nu=1),  # with bs_mul = linear\n",
    "#     \"clipped_SSTM, 1e-3, 1, norm, 1, bs_mul = linear(400), nu=1\",\n",
    "# ]\n",
    "\n",
    "opt_names = []\n",
    "\n",
    "for i in range(5):\n",
    "#     for i, (lr, batch_mul_step_count) in enumerate(itertools.product(lrs, batch_mul_step_count_list)):\n",
    "#         opt_names.append(\"SGD, {lr}, {momentum}, bs_mul = 1\".format(lr=lr, momentum=momentum))\n",
    "#         opt_names.append(\"clipped_SGD, {lr}, {momentum}, coordinate_wise, 0.1, bs_mul = 1\".format(lr=lr, momentum=momentum))\n",
    "#         opt_names.append(\"clipped_SSTM, {lr}, {momentum}, coordinate_wise, 0.1, bs_mul = 1\".format(lr=lr, momentum=momentum)))\n",
    "#         opt_names.append(\"clipped_SSTM, {lr}, 1, norm, 1, bs_mul = linear ({batch_mul_step_count}), nu=1\".format(\n",
    "#             lr=lr, batch_mul_step_count=batch_mul_step_count))\n",
    "\n",
    "    opt_names += [ # normal opts names\n",
    "        \"Adam, 5e-4, bs_mul = 1\",\n",
    "        \"SGD, 3e-4, 0.99, bs_mul = 1\",\n",
    "        \"clipped_SGD, 1e-3, 0.95, coordinate_wise, 0.1, bs_mul = 1\",\n",
    "        \"clipped_SSTM, 3e-3, 1, norm, 1, bs_mul = 4, nu=1\",\n",
    "        \"clipped_SSTM, 1e-2, 1, norm, 1, bs_mul = linear (100), nu=1\",\n",
    "    ]\n",
    "\n",
    "# bs_muls = [\n",
    "#     8,\n",
    "# #     1, 1, 1, 8, \"linear\"\n",
    "# ] * 5 * shift_mul\n",
    "\n",
    "lr_decays = [\n",
    "#     0.999, 0.999, 0.999, 0.999, 0.999\n",
    "    1,\n",
    "] * 5 * 5\n",
    "\n",
    "# batch_mul_step_counts = [\n",
    "#     400,\n",
    "# ] * 5 * shift_mul\n",
    "\n",
    "hist = []\n",
    "\n",
    "for (net, optimizer, opt_name, bs_mul, lr_decay, batch_mul_step_count) in zip(nets, opts, opt_names, bs_muls, lr_decays, batch_mul_step_counts):\n",
    "    hist.append({\n",
    "        \"task_name\": \"BERT on MRPC\",\n",
    "        \"name\": opt_name, \n",
    "        \"bs_mul\": bs_mul,\n",
    "        \"lr_decay\": lr_decay,\n",
    "        \"train_loss\": [], \"train_x\": [], \n",
    "        \"val_loss\": [], \"val_x\": [], \n",
    "        \"train_acc_top_1\": [], \"train_acc_top_5\": [], \n",
    "        \"val_acc_top_1\": [], \"val_acc_top_5\": [], \n",
    "        \"norm_diffs\": [],\n",
    "        \"epochs_x\": [],\n",
    "        \"norm_diffs_x\": [],\n",
    "        \"total_steps\": 0,\n",
    "        \"prev_val_eval_step\": 0,\n",
    "        \"prev_grad_norms_eval_step\": 0,\n",
    "        \"batch_end\": True,\n",
    "        \"batch_mul_step_count\": batch_mul_step_count\n",
    "    })"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Norm diffs functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy\n",
    "from scipy import stats\n",
    "from tqdm.notebook import tqdm\n",
    "\n",
    "def get_batch_grad(model):\n",
    "    gr=[]\n",
    "    for i in model.parameters():\n",
    "        if i.requires_grad:\n",
    "            gr.append(i.grad.view(-1))\n",
    "    return torch.cat(gr)\n",
    "\n",
    "\n",
    "def get_loss(model, criterion, batch):\n",
    "    inputs, labels = batch[0].to(device), batch[1].to(device)\n",
    "    outputs = model(inputs)\n",
    "    loss = criterion(outputs, labels)\n",
    "    return loss\n",
    "\n",
    "\n",
    "def compute_full_grad(model, optimizer, criterion, dataloader_for_full_grad):\n",
    "    fully_grad = []\n",
    "    optimizer.zero_grad()\n",
    "\n",
    "    print(\"Computing full gradient\")\n",
    "    with tqdm(total=len(dataloader_for_full_grad)) as pbar:\n",
    "        for step, batch in enumerate(dataloader_for_full_grad):\n",
    "            loss = get_loss(model, criterion, batch)\n",
    "            loss.backward()\n",
    "\n",
    "            if fully_grad != []:\n",
    "                fully_grad = fully_grad + get_batch_grad(model) * \\\n",
    "                    dataloader_for_full_grad.batch_size\n",
    "            else:\n",
    "                fully_grad = get_batch_grad(model) * dataloader_for_full_grad.batch_size\n",
    "            optimizer.zero_grad()\n",
    "\n",
    "            pbar.update(1)\n",
    "\n",
    "    return fully_grad / (step * dataloader_for_full_grad.batch_size)\n",
    "\n",
    "\n",
    "def compute_norm_diffs(model, optimizer, criterion, dataloader_for_full_grad, dataloader, full_grad=None, repeats=1):\n",
    "    if full_grad is None:\n",
    "        full_grad = compute_full_grad(model, optimizer, criterion, dataloader_for_full_grad)\n",
    "    mini_norms = []\n",
    "    optimizer.zero_grad()\n",
    "\n",
    "    print(\"Computing norm diffs\")\n",
    "    with tqdm(total=repeats * len(dataloader)) as pbar:\n",
    "        for _ in range(repeats):\n",
    "            for step, batch in enumerate(dataloader):\n",
    "                loss = get_loss(model, criterion, batch)\n",
    "                loss.backward()\n",
    "\n",
    "                mini_norms.append((get_batch_grad(model) - full_grad).norm().item())\n",
    "                optimizer.zero_grad()\n",
    "\n",
    "                pbar.update(1)\n",
    "\n",
    "    return np.array(mini_norms)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from copy import deepcopy\n",
    "import numpy as np\n",
    "\n",
    "def get_loss(model, criterion, batch):\n",
    "    b_input_ids = batch[0].to(device)\n",
    "    b_input_mask = batch[1].to(device)\n",
    "    b_labels = batch[2].to(device)\n",
    "    outputs = model(\n",
    "        b_input_ids, \n",
    "        token_type_ids=None, \n",
    "        attention_mask=b_input_mask, \n",
    "        labels=b_labels\n",
    "    )\n",
    "\n",
    "    loss=outputs[0]\n",
    "    return loss\n",
    "\n",
    "def recursive_to(param, device):\n",
    "    # Not sure if there are any global tensors in the state dict\n",
    "    if isinstance(param, torch.Tensor):\n",
    "        param.data = param.data.to(device)\n",
    "        if param._grad is not None:\n",
    "            param._grad.data = param._grad.data.to(device)\n",
    "    elif isinstance(param, dict):\n",
    "        for subparam in param.values():\n",
    "            recursive_to(subparam, device)\n",
    "    elif isinstance(param, list):\n",
    "        for subparam in param:\n",
    "            recursive_to(subparam, device)\n",
    "\n",
    "def optimizer_to(optim, device):\n",
    "    for param_group in optim.param_groups:\n",
    "        for param in param_group.values():\n",
    "            # Not sure there are any global tensors in the state dict\n",
    "            recursive_to(param, device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x864 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Finished Training\n"
     ]
    }
   ],
   "source": [
    "from IPython import display\n",
    "\n",
    "# batch_mul_step_count = 50\n",
    "calc_norm_diffs = False\n",
    "\n",
    "\n",
    "for epoch in range(15):\n",
    "    for (net, optimizer, net_hist) in zip(nets, opts, hist):\n",
    "        net.to(device)\n",
    "        optimizer_to(optimizer, device)\n",
    "\n",
    "        total_steps = net_hist[\"total_steps\"]\n",
    "        # effective batch size will be batch_size * bs_mul\n",
    "        bs_mul = net_hist[\"bs_mul\"]\n",
    "        # learning rate decay rate\n",
    "        lr_decay = net_hist[\"lr_decay\"]\n",
    "        batch_mul_step_count = net_hist[\"batch_mul_step_count\"]\n",
    "            \n",
    "        if net_hist[\"bs_mul\"] == \"linear\":\n",
    "            if not (\"bs_mul_value\" in net_hist):\n",
    "                net_hist[\"bs_mul_value\"] = 1\n",
    "\n",
    "            bs_mul = net_hist[\"bs_mul_value\"]\n",
    "            \n",
    "        net_hist[\"epochs_x\"].append(total_steps)\n",
    "\n",
    "        running_loss = 0.0\n",
    "        for i, data in enumerate(train_loader, 0):\n",
    "            # compute norm diffs distribution on train dataset\n",
    "            prev_grad_norms_eval_step = net_hist[\"prev_grad_norms_eval_step\"]\n",
    "            if calc_norm_diffs and ((total_steps - prev_grad_norms_eval_step) > 500 or len(net_hist[\"norm_diffs\"]) == 0) and net_hist[\"batch_end\"]:\n",
    "                net_hist[\"prev_grad_norms_eval_step\"] = total_steps\n",
    "                net.eval()\n",
    "\n",
    "                norm_diffs = compute_norm_diffs(\n",
    "                    net, optimizer, None, \n",
    "                    train_loader_for_full_grad, train_loader, repeats=5\n",
    "                )\n",
    "\n",
    "                net_hist[\"norm_diffs\"].append(norm_diffs)\n",
    "                net_hist[\"norm_diffs_x\"].append(total_steps)\n",
    "                net.train()\n",
    "\n",
    "\n",
    "            # evaluate on validation dataset\n",
    "            prev_val_eval_step = net_hist[\"prev_val_eval_step\"]\n",
    "            if ((total_steps - prev_val_eval_step) > 20 or len(net_hist[\"val_loss\"]) == 0) and net_hist[\"batch_end\"]:\n",
    "                net_hist[\"prev_val_eval_step\"] = total_steps\n",
    "\n",
    "                net.eval()\n",
    "\n",
    "                val_losses = []\n",
    "                val_accs = []\n",
    "                \n",
    "                with torch.no_grad():\n",
    "                    for step, val_data in enumerate(val_loader):\n",
    "\n",
    "                        inputs, masks, labels = val_data[0].to(device), val_data[1].to(device), val_data[2].to(device)\n",
    "\n",
    "                        outputs = net(\n",
    "                            inputs, \n",
    "                            token_type_ids=None, \n",
    "                            attention_mask=masks, \n",
    "                            labels=labels\n",
    "                        )\n",
    "                        loss = outputs[0]\n",
    "\n",
    "                        val_losses.append(loss.detach().cpu().item())\n",
    "\n",
    "                        acc = accuracy(outputs.logits, labels.data, topk=(1,))\n",
    "                        val_accs.append(acc[0].detach().cpu().item())\n",
    "\n",
    "                net_hist[\"val_loss\"].append(np.mean(val_losses))\n",
    "                net_hist[\"val_x\"].append(total_steps)\n",
    "\n",
    "                net_hist[\"val_acc_top_1\"].append(np.mean(val_accs))\n",
    "\n",
    "                net.train()\n",
    "\n",
    "            net_hist[\"batch_end\"] = False\n",
    "            ######################################################################\n",
    "            if lr_decay < 1:\n",
    "                for g in optimizer.param_groups:\n",
    "                    g['lr'] = g['lr'] * lr_decay\n",
    "            ######################################################################\n",
    "            \n",
    "            # train\n",
    "            inputs, masks, labels = data[0].to(device), data[1].to(device), data[2].to(device)\n",
    "\n",
    "            outputs = net(\n",
    "                inputs, \n",
    "                token_type_ids=None, \n",
    "                attention_mask=masks, \n",
    "                labels=labels\n",
    "            )\n",
    "            loss = outputs[0] / bs_mul\n",
    "            loss.backward()\n",
    "\n",
    "            if total_steps % bs_mul == bs_mul - 1:\n",
    "                optimizer.step()\n",
    "                optimizer.zero_grad()\n",
    "                net_hist[\"batch_end\"] = True\n",
    "\n",
    "            net_hist[\"train_loss\"].append(loss.detach().cpu().item() * bs_mul)\n",
    "            net_hist[\"train_x\"].append(total_steps)\n",
    "\n",
    "            if total_steps % bs_mul == bs_mul - 1:\n",
    "                if net_hist[\"bs_mul\"] == \"linear\":\n",
    "                    net_hist[\"bs_mul_value\"] = int(int(total_steps) / batch_mul_step_count) + 1 \n",
    "                    bs_mul = net_hist[\"bs_mul_value\"]\n",
    "\n",
    "            top_1 = accuracy(outputs.logits, labels.data, topk=(1,))\n",
    "            net_hist[\"train_acc_top_1\"].append(top_1[0].detach().cpu().item())\n",
    "\n",
    "            if total_steps % 10 == 0:\n",
    "                display.clear_output(wait=True)\n",
    "                \n",
    "                grouped_hist = group_uniques_full(hist, [\"train_loss\", \"val_loss\", \"val_acc_top_1\", \"train_acc_top_1\"])\n",
    "\n",
    "                fig = plt.figure(figsize=(15, 8 + 2 * ((len(grouped_hist) + 2) // 3)))\n",
    "                gs = GridSpec(4 + 2 * ((len(grouped_hist) + 2) // 3),3, figure=fig)\n",
    "\n",
    "                ax1 = fig.add_subplot(gs[0:4,:2])\n",
    "                ax2 = fig.add_subplot(gs[0:2,2])\n",
    "                ax3 = fig.add_subplot(gs[2:4,2])\n",
    "\n",
    "                ax1 = make_loss_plot(ax1, grouped_hist, eps=0.01, make_val=False, alpha=0.9)\n",
    "                ax2 = make_accuracy_plot(ax2, grouped_hist, eps=0.01, make_train=True, make_val=False, top_k=1, alpha=0.9)\n",
    "                ax3 = make_accuracy_plot(ax3, grouped_hist, eps=0.01, make_train=False, make_val=True, top_k=1, alpha=0.9)\n",
    "                \n",
    "                if calc_norm_diffs:\n",
    "                    draw_norm_hists_for_different_models(fig, gs[4:,:], grouped_hist, bins_n=100, draw_normal=True)\n",
    "\n",
    "                gs.tight_layout(fig)\n",
    "                plt.draw()\n",
    "                plt.show()\n",
    "\n",
    "            total_steps += 1\n",
    "        net_hist[\"total_steps\"] = total_steps\n",
    "        net.to(\"cpu\")\n",
    "        optimizer_to(optimizer, \"cpu\")\n",
    "\n",
    "print('Finished Training')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Save models / train data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "\n",
    "now = datetime.now()\n",
    "now = str(now)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_net_weights = False\n",
    "\n",
    "\n",
    "if save_net_weights:\n",
    "    for i in range(len(nets)):\n",
    "        if \"/\" in hist[i][\"name\"]:\n",
    "            hist[i][\"name\"] = \":\".join(hist[i][\"name\"].split(\"/\"))\n",
    "        PATH = './models/bert_{}.pth'.format(\n",
    "            \"_\".join(now.split(\" \")[0:1] + now.split(\" \")[1].split(\".\")[0].split(\":\") + hist[i][\"name\"].split())\n",
    "        )\n",
    "        torch.save(nets[0].state_dict(), PATH)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from copy import deepcopy\n",
    "\n",
    "hist3 = deepcopy(hist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "for h in hist3:\n",
    "    for key in h:\n",
    "        if isinstance(h[key], list) and  len(h[key])> 0 :\n",
    "            if isinstance(h[key][0], torch.Tensor):\n",
    "                for i in range(len(h[key])):\n",
    "                    h[key][i] = h[key][i].numpy()\n",
    "            if isinstance(h[key][0], np.ndarray):\n",
    "                for i in range(len(h[key])):\n",
    "                    if len(h[key][i]) == 1:\n",
    "                        h[key][i] = np.asscalar(h[key][i])\n",
    "                    else:\n",
    "                        h[key][i] = [str(x) for x in h[key][i]]\n",
    "            for i in range(len(h[key])):\n",
    "                if not isinstance(h[key][i], list):\n",
    "                    h[key][i] = str(h[key][i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import os\n",
    "\n",
    "filename = './results/{}/mrpc/bert_mrpc_{}_train_hist.json'.format(data_folder_name,\"_\".join(now.split(\" \")[0:1] + now.split(\" \")[1].split(\".\")[0].split(\":\")))\n",
    "          \n",
    "os.makedirs(os.path.dirname(filename), exist_ok=True)\n",
    "with open(filename, 'w') as fout:\n",
    "    json.dump(hist3 , fout)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Averaged plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "hists_names_list = [\n",
    "    \"bert_mrpc_2021-09-22_21_31_00_train_hist\", # combined (ds-21, 21, 100, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-22_16_27_29_train_hist\", # combined (ds-20, 20, 95, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-22_07_51_28_train_hist\", # combined (ds-19, 19, 90, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-22_03_35_02_train_hist\", # combined (ds-18, 18, 85, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-21_23_10_41_train_hist\", # combined (ds-17, 17, 80, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-21_17_32_42_train_hist\", # combined (ds-16, 16, 75, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-21_11_22_31_train_hist\", # combined (ds-15, 15, 70, 15 epochs) + norm diffs\n",
    "    \"bert_mrpc_2021-09-20_22_03_29_train_hist\", # combined (ds-14, 14, 65, 15 epochs) + norm diffs\n",
    "    \"bert_mrpc_2021-09-20_09_24_37_train_hist\", # combined (ds-13, 13, 60, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-20_03_45_56_train_hist\", # combined (ds-12, 12, 55, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-19_21_34_24_train_hist\", # combined (ds-11, 11, 50, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-19_17_12_13_train_hist\", # combined (ds-10, 10, 45, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-19_11_55_58_train_hist\", # combined (ds-9, 9, 45, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-19_00_27_21_train_hist\", # combined (ds-8, 8, 40, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-18_17_06_17_train_hist\", # combined (ds-7, 7, 35, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-18_02_32_08_train_hist\", # combined (ds-6, 6, 30, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-17_21_13_59_train_hist\", # combined (ds-5, 5, 25, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-17_15_39_06_train_hist\", # combined (ds-4, 4, 20, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-17_10_20_20_train_hist\", # combined (ds-3, 3, 15, 15 epochs)\n",
    "    \"bert_mrpc_2021-09-17_00_34_18_train_hist\", # combined (ds-2, 2, 10, 15 epochs)\n",
    "]\n",
    "\n",
    "# hists_names_list = hists_names_list[::-1]\n",
    "\n",
    "\n",
    "hist = hist_from_json(load_hist_jsons(hists_names_list, path=\"./results/glue_data/mrpc/\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Adam, 5e-4, bs_mul = 1'"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hist[0][\"name\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Repeats_1 = 100.0, Name = Adam, 5e-4, bs_mul = 1\n",
      "Repeats_1 = 100.0, Name = SGD, 3e-4, 0.99, bs_mul = 1\n",
      "Repeats_1 = 100.0, Name = clipped_SGD, 1e-3, 0.95, coordinate_wise, 0.1, bs_mul = 1\n",
      "Repeats_1 = 100.0, Name = clipped_SSTM, 3e-3, 1, norm, 1, bs_mul = 4, nu=1\n",
      "Repeats_1 = 100.0, Name = clipped_SSTM, 1e-2, 1, norm, 1, bs_mul = linear (100), nu=1\n"
     ]
    }
   ],
   "source": [
    "grouped_hist = group_uniques_full(\n",
    "    hist, \n",
    "    [\"train_loss\", \"val_loss\", \"train_acc_top_1\", \"val_acc_top_1\"], \n",
    "    verbose=True,\n",
    "    group_norm_diffs=True\n",
    ")\n",
    "\n",
    "# grouped_hist = grouped_hist[0:4]\n",
    "\n",
    "def key(h, alpha=0.01):\n",
    "    return smooth(h[\"train_loss\"], alpha)[-1]\n",
    "\n",
    "grouped_hist = sorted(grouped_hist, key=key)\n",
    "\n",
    "def select_top(grouped_hist, algos):\n",
    "    res = dict([(name, None) for name in algos])\n",
    "    for name in algos:\n",
    "        for hist in grouped_hist:\n",
    "            if name == hist[\"name\"].split(\",\")[0]:\n",
    "                res[name] = hist\n",
    "                break\n",
    "    return list(res.values())\n",
    "\n",
    "# grouped_hist = select_top(grouped_hist, [\"clipped_SSTM\", \"clipped_SGD\", \"SGD\", \"Adam\"])\n",
    "\n",
    "grouped_hist = grouped_hist[:15]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x864 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython import display\n",
    "\n",
    "fig = plt.figure(figsize=(15,12))\n",
    "gs = GridSpec(3,2, figure=fig)\n",
    "\n",
    "\n",
    "display.clear_output(wait=True)\n",
    "\n",
    "ax1 = fig.add_subplot(gs[0,0])\n",
    "ax2 = fig.add_subplot(gs[0,1])\n",
    "ax3 = fig.add_subplot(gs[1,0])\n",
    "ax4 = fig.add_subplot(gs[1,1])\n",
    "\n",
    "ax1 = make_loss_plot(ax1, grouped_hist, eps=0.01, make_val=False, alpha=0.9, starting_epoch=0)\n",
    "ax2 = make_loss_plot(ax2, grouped_hist, eps=0.01, make_train=False, alpha=0.9, starting_epoch=0)\n",
    "ax3 = make_accuracy_plot(ax3, grouped_hist, eps=0.01, make_train=True, make_val=False, top_k=1, alpha=0.9, starting_epoch=1)\n",
    "ax4 = make_accuracy_plot(ax4, grouped_hist, eps=0.01, make_train=False, make_val=True, top_k=1, alpha=0.9, starting_epoch=1)\n",
    "\n",
    "grouped_hist_for_distrs = [grouped_hist[0], grouped_hist[1], grouped_hist[3]]\n",
    "\n",
    "draw_norm_hists_for_different_models(fig, gs[2:,:], grouped_hist_for_distrs, bins_n=100, draw_normal=True, number=0)\n",
    "\n",
    "gs.tight_layout(fig)\n",
    "plt.draw()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}