Copy_of_S+P_Week_3_Exercise_Question.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "Copy of S+P Week 3 Exercise Question.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/RhuthuHegde/Tensorflow-coursera/blob/master/Copy_of_S%2BP_Week_3_Exercise_Question.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zX4Kg8DUTKWO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "D1J15Vh_1Jih",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "933166ec-ac68-47a0-8eaa-b1fb0205cc5e"
      },
      "source": [
        "!pip install tensorflow==2.0.0\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting tensorflow==2.0.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/46/0f/7bd55361168bb32796b360ad15a25de6966c9c1beb58a8e30c01c8279862/tensorflow-2.0.0-cp36-cp36m-manylinux2010_x86_64.whl (86.3MB)\n",
            "\u001b[K     |████████████████████████████████| 86.3MB 53kB/s \n",
            "\u001b[?25hRequirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (3.3.0)\n",
            "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (1.1.0)\n",
            "Collecting tensorboard<2.1.0,>=2.0.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/76/54/99b9d5d52d5cb732f099baaaf7740403e83fe6b0cedde940fabd2b13d75a/tensorboard-2.0.2-py3-none-any.whl (3.8MB)\n",
            "\u001b[K     |████████████████████████████████| 3.8MB 49.0MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy<2.0,>=1.16.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (1.18.5)\n",
            "Collecting tensorflow-estimator<2.1.0,>=2.0.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/fc/08/8b927337b7019c374719145d1dceba21a8bb909b93b1ad6f8fb7d22c1ca1/tensorflow_estimator-2.0.1-py2.py3-none-any.whl (449kB)\n",
            "\u001b[K     |████████████████████████████████| 450kB 41.7MB/s \n",
            "\u001b[?25hRequirement already satisfied: absl-py>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (0.9.0)\n",
            "Requirement already satisfied: protobuf>=3.6.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (3.12.4)\n",
            "Requirement already satisfied: google-pasta>=0.1.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (0.2.0)\n",
            "Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (1.15.0)\n",
            "Requirement already satisfied: grpcio>=1.8.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (1.31.0)\n",
            "Requirement already satisfied: wheel>=0.26 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (0.34.2)\n",
            "Requirement already satisfied: wrapt>=1.11.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (1.12.1)\n",
            "Collecting keras-applications>=1.0.8\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/71/e3/19762fdfc62877ae9102edf6342d71b28fbfd9dea3d2f96a882ce099b03f/Keras_Applications-1.0.8-py3-none-any.whl (50kB)\n",
            "\u001b[K     |████████████████████████████████| 51kB 6.9MB/s \n",
            "\u001b[?25hRequirement already satisfied: astor>=0.6.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (0.8.1)\n",
            "Collecting gast==0.2.2\n",
            "  Downloading https://files.pythonhosted.org/packages/4e/35/11749bf99b2d4e3cceb4d55ca22590b0d7c2c62b9de38ac4a4a7f4687421/gast-0.2.2.tar.gz\n",
            "Requirement already satisfied: keras-preprocessing>=1.0.5 in /usr/local/lib/python3.6/dist-packages (from tensorflow==2.0.0) (1.1.2)\n",
            "Requirement already satisfied: werkzeug>=0.11.15 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (1.0.1)\n",
            "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (0.4.1)\n",
            "Requirement already satisfied: google-auth<2,>=1.6.3 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (1.17.2)\n",
            "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (2.23.0)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (3.2.2)\n",
            "Requirement already satisfied: setuptools>=41.0.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (49.2.0)\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.6/dist-packages (from keras-applications>=1.0.8->tensorflow==2.0.0) (2.10.0)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (1.3.0)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4; python_version >= \"3\" in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (4.6)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (0.2.8)\n",
            "Requirement already satisfied: cachetools<5.0,>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (4.1.1)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (2.10)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (3.0.4)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (1.24.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3,>=2.21.0->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (2020.6.20)\n",
            "Requirement already satisfied: importlib-metadata; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from markdown>=2.6.8->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (1.7.0)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.6/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (3.1.0)\n",
            "Requirement already satisfied: pyasn1>=0.1.3 in /usr/local/lib/python3.6/dist-packages (from rsa<5,>=3.1.4; python_version >= \"3\"->google-auth<2,>=1.6.3->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (0.4.8)\n",
            "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata; python_version < \"3.8\"->markdown>=2.6.8->tensorboard<2.1.0,>=2.0.0->tensorflow==2.0.0) (3.1.0)\n",
            "Building wheels for collected packages: gast\n",
            "  Building wheel for gast (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for gast: filename=gast-0.2.2-cp36-none-any.whl size=7540 sha256=8478fc5d8669651d8a13666467bf6a4dd6c7edfc5fbe07a47b79cb6e8141e869\n",
            "  Stored in directory: /root/.cache/pip/wheels/5c/2e/7e/a1d4d4fcebe6c381f378ce7743a3ced3699feb89bcfbdadadd\n",
            "Successfully built gast\n",
            "\u001b[31mERROR: tensorflow-probability 0.11.0 has requirement gast>=0.3.2, but you'll have gast 0.2.2 which is incompatible.\u001b[0m\n",
            "Installing collected packages: tensorboard, tensorflow-estimator, keras-applications, gast, tensorflow\n",
            "  Found existing installation: tensorboard 2.3.0\n",
            "    Uninstalling tensorboard-2.3.0:\n",
            "      Successfully uninstalled tensorboard-2.3.0\n",
            "  Found existing installation: tensorflow-estimator 2.3.0\n",
            "    Uninstalling tensorflow-estimator-2.3.0:\n",
            "      Successfully uninstalled tensorflow-estimator-2.3.0\n",
            "  Found existing installation: gast 0.3.3\n",
            "    Uninstalling gast-0.3.3:\n",
            "      Successfully uninstalled gast-0.3.3\n",
            "  Found existing installation: tensorflow 2.3.0\n",
            "    Uninstalling tensorflow-2.3.0:\n",
            "      Successfully uninstalled tensorflow-2.3.0\n",
            "Successfully installed gast-0.2.2 keras-applications-1.0.8 tensorboard-2.0.2 tensorflow-2.0.0 tensorflow-estimator-2.0.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "BOjujz601HcS",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "09780c05-cbdf-43a5-f9ae-4cff16c93e1d"
      },
      "source": [
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "print(tf.__version__)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2.0.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "Zswl7jRtGzkk",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "outputId": "f04cc60e-cc3d-4ce2-8e98-c6181ff2b810"
      },
      "source": [
        "def plot_series(time, series, format=\"-\", start=0, end=None):\n",
        "    plt.plot(time[start:end], series[start:end], format)\n",
        "    plt.xlabel(\"Time\")\n",
        "    plt.ylabel(\"Value\")\n",
        "    plt.grid(False)\n",
        "\n",
        "def trend(time, slope=0):\n",
        "    return slope * time\n",
        "\n",
        "def seasonal_pattern(season_time):\n",
        "    \"\"\"Just an arbitrary pattern, you can change it if you wish\"\"\"\n",
        "    return np.where(season_time < 0.1,\n",
        "                    np.cos(season_time * 6 * np.pi),\n",
        "                    2 / np.exp(9 * season_time))\n",
        "\n",
        "def seasonality(time, period, amplitude=1, phase=0):\n",
        "    \"\"\"Repeats the same pattern at each period\"\"\"\n",
        "    season_time = ((time + phase) % period) / period\n",
        "    return amplitude * seasonal_pattern(season_time)\n",
        "\n",
        "def noise(time, noise_level=1, seed=None):\n",
        "    rnd = np.random.RandomState(seed)\n",
        "    return rnd.randn(len(time)) * noise_level\n",
        "\n",
        "time = np.arange(10 * 365 + 1, dtype=\"float32\")\n",
        "baseline = 10\n",
        "series = trend(time, 0.1)  \n",
        "baseline = 10\n",
        "amplitude = 40\n",
        "slope = 0.005\n",
        "noise_level = 3\n",
        "\n",
        "# Create the series\n",
        "series = baseline + trend(time, slope) + seasonality(time, period=365, amplitude=amplitude)\n",
        "# Update with noise\n",
        "series += noise(time, noise_level, seed=51)\n",
        "\n",
        "split_time = 3000\n",
        "time_train = time[:split_time]\n",
        "x_train = series[:split_time]\n",
        "time_valid = time[split_time:]\n",
        "x_valid = series[split_time:]\n",
        "\n",
        "window_size = 20\n",
        "batch_size = 32\n",
        "shuffle_buffer_size = 1000\n",
        "\n",
        "plot_series(time, series)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "4sTTIOCbyShY",
        "colab": {}
      },
      "source": [
        "def windowed_dataset(series, window_size, batch_size, shuffle_buffer):\n",
        "  dataset = tf.data.Dataset.from_tensor_slices(series)\n",
        "  dataset = dataset.window(window_size + 1, shift=1, drop_remainder=True)\n",
        "  dataset = dataset.flat_map(lambda window: window.batch(window_size + 1))\n",
        "  dataset = dataset.shuffle(shuffle_buffer).map(lambda window: (window[:-1], window[-1]))\n",
        "  dataset = dataset.batch(batch_size).prefetch(1)\n",
        "  return dataset"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "A1Hl39rklkLm",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "dd36cf47-5a60-4eb6-d72e-36c2047f2c57"
      },
      "source": [
        "tf.keras.backend.clear_session()\n",
        "tf.random.set_seed(51)\n",
        "np.random.seed(51)\n",
        "\n",
        "tf.keras.backend.clear_session()\n",
        "dataset = windowed_dataset(x_train, window_size, batch_size, shuffle_buffer_size)\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "  tf.keras.layers.Lambda(lambda x: tf.expand_dims(x,axis=-1),input_shape=[None]),\n",
        "  tf.keras.layers.SimpleRNN(40,return_sequences=True),\n",
        "  tf.keras.layers.SimpleRNN(40),\n",
        "  tf.keras.layers.Dense(1),\n",
        "  # YOUR CODE HERE\n",
        "  tf.keras.layers.Lambda(lambda x: x*100.0)\n",
        "])\n",
        "\n",
        "lr_schedule = tf.keras.callbacks.LearningRateScheduler(\n",
        "    lambda epoch: 1e-8 * 10**(epoch / 20))\n",
        "optimizer = tf.keras.optimizers.SGD(lr=1e-8, momentum=0.9)\n",
        "model.compile(loss=tf.keras.losses.Huber(),\n",
        "              optimizer=optimizer,\n",
        "              metrics=[\"mae\"])\n",
        "history = model.fit(dataset, epochs=100, callbacks=[lr_schedule])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "94/94 [==============================] - 2s 26ms/step - loss: 169.3279 - mae: 169.9036\n",
            "Epoch 2/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 139.4066 - mae: 139.9221\n",
            "Epoch 3/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 104.8306 - mae: 105.3155\n",
            "Epoch 4/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 68.2253 - mae: 68.7019\n",
            "Epoch 5/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 33.0069 - mae: 33.4813\n",
            "Epoch 6/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 11.2019 - mae: 11.6840\n",
            "Epoch 7/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 8.1500 - mae: 8.6331\n",
            "Epoch 8/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 7.5348 - mae: 8.0145\n",
            "Epoch 9/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 7.4230 - mae: 7.9063\n",
            "Epoch 10/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 7.3544 - mae: 7.8350\n",
            "Epoch 11/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 7.2857 - mae: 7.7647\n",
            "Epoch 12/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 7.1967 - mae: 7.6871\n",
            "Epoch 13/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 7.1362 - mae: 7.6137\n",
            "Epoch 14/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 7.0393 - mae: 7.5130\n",
            "Epoch 15/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 6.9391 - mae: 7.4161\n",
            "Epoch 16/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 6.8486 - mae: 7.3264\n",
            "Epoch 17/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 6.7454 - mae: 7.2199\n",
            "Epoch 18/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 6.6106 - mae: 7.0842\n",
            "Epoch 19/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 6.4728 - mae: 6.9500\n",
            "Epoch 20/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 6.3528 - mae: 6.8281\n",
            "Epoch 21/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 6.2114 - mae: 6.6881\n",
            "Epoch 22/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 6.0649 - mae: 6.5414\n",
            "Epoch 23/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 5.9151 - mae: 6.3897\n",
            "Epoch 24/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 5.7085 - mae: 6.1845\n",
            "Epoch 25/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 5.5161 - mae: 5.9869\n",
            "Epoch 26/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 5.3075 - mae: 5.7800\n",
            "Epoch 27/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 5.1432 - mae: 5.6165\n",
            "Epoch 28/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.9795 - mae: 5.4556\n",
            "Epoch 29/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.7403 - mae: 5.2149\n",
            "Epoch 30/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 4.5689 - mae: 5.0422\n",
            "Epoch 31/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 4.3727 - mae: 4.8414\n",
            "Epoch 32/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.2351 - mae: 4.7060\n",
            "Epoch 33/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.0316 - mae: 4.5099\n",
            "Epoch 34/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.9048 - mae: 4.3753\n",
            "Epoch 35/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.7408 - mae: 4.2069\n",
            "Epoch 36/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.5608 - mae: 4.0415\n",
            "Epoch 37/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.5215 - mae: 3.9873\n",
            "Epoch 38/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.3575 - mae: 3.8237\n",
            "Epoch 39/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 3.3180 - mae: 3.7856\n",
            "Epoch 40/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.2884 - mae: 3.7522\n",
            "Epoch 41/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.2788 - mae: 3.7422\n",
            "Epoch 42/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.3861 - mae: 3.8535\n",
            "Epoch 43/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.2871 - mae: 3.7511\n",
            "Epoch 44/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.1497 - mae: 3.6184\n",
            "Epoch 45/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.3126 - mae: 3.7796\n",
            "Epoch 46/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.3090 - mae: 3.7758\n",
            "Epoch 47/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.1078 - mae: 3.5839\n",
            "Epoch 48/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.1993 - mae: 3.6649\n",
            "Epoch 49/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.0775 - mae: 3.5427\n",
            "Epoch 50/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 3.2555 - mae: 3.7209\n",
            "Epoch 51/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 3.1373 - mae: 3.5964\n",
            "Epoch 52/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.0021 - mae: 3.4662\n",
            "Epoch 53/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.1723 - mae: 3.6447\n",
            "Epoch 54/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.1456 - mae: 3.6539\n",
            "Epoch 55/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 2.9448 - mae: 3.4099\n",
            "Epoch 56/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.2395 - mae: 3.7101\n",
            "Epoch 57/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 3.3605 - mae: 3.8273\n",
            "Epoch 58/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.0693 - mae: 3.5331\n",
            "Epoch 59/100\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 3.0372 - mae: 3.5024\n",
            "Epoch 60/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.2216 - mae: 3.6913\n",
            "Epoch 61/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.4604 - mae: 3.9324\n",
            "Epoch 62/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.1917 - mae: 3.6567\n",
            "Epoch 63/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.3317 - mae: 3.8003\n",
            "Epoch 64/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.2966 - mae: 3.7642\n",
            "Epoch 65/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.4612 - mae: 3.9329\n",
            "Epoch 66/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.1161 - mae: 4.5902\n",
            "Epoch 67/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.2235 - mae: 3.6908\n",
            "Epoch 68/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 3.9872 - mae: 4.4586\n",
            "Epoch 69/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.3461 - mae: 3.8157\n",
            "Epoch 70/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.4019 - mae: 4.8855\n",
            "Epoch 71/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 3.8011 - mae: 4.2698\n",
            "Epoch 72/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 3.9509 - mae: 4.4246\n",
            "Epoch 73/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.0758 - mae: 4.5513\n",
            "Epoch 74/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 6.1339 - mae: 6.6163\n",
            "Epoch 75/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 5.6665 - mae: 6.1505\n",
            "Epoch 76/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 6.5126 - mae: 6.9938\n",
            "Epoch 77/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 9.2148 - mae: 9.7008\n",
            "Epoch 78/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 8.1407 - mae: 8.6427\n",
            "Epoch 79/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 13.9017 - mae: 14.3846\n",
            "Epoch 80/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 6.1096 - mae: 6.5925\n",
            "Epoch 81/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 5.8212 - mae: 6.2978\n",
            "Epoch 82/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 7.1824 - mae: 7.6617\n",
            "Epoch 83/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 13.1714 - mae: 13.6519\n",
            "Epoch 84/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 9.3481 - mae: 9.8443\n",
            "Epoch 85/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.8470 - mae: 5.3235\n",
            "Epoch 86/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 9.1376 - mae: 9.6191\n",
            "Epoch 87/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 6.9362 - mae: 7.4188\n",
            "Epoch 88/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 5.3100 - mae: 5.7987\n",
            "Epoch 89/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 6.4018 - mae: 6.8819\n",
            "Epoch 90/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 5.5911 - mae: 6.0735\n",
            "Epoch 91/100\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 4.8445 - mae: 5.3216\n",
            "Epoch 92/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 4.5752 - mae: 5.0512\n",
            "Epoch 93/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 5.3933 - mae: 5.8675\n",
            "Epoch 94/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 8.1495 - mae: 8.6303\n",
            "Epoch 95/100\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 10.5360 - mae: 11.0227\n",
            "Epoch 96/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 9.5250 - mae: 10.0027\n",
            "Epoch 97/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 7.6468 - mae: 8.1317\n",
            "Epoch 98/100\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 11.2314 - mae: 11.7656\n",
            "Epoch 99/100\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 115.8048 - mae: 116.2129\n",
            "Epoch 100/100\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 120.5905 - mae: 121.1368\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "AkBsrsXMzoWR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 290
        },
        "outputId": "ff079ab2-2356-450d-b01b-cd31dc5a14c7"
      },
      "source": [
        "plt.semilogx(history.history[\"lr\"], history.history[\"loss\"])\n",
        "plt.axis([1e-8, 1e-4, 0, 30])\n",
        "\n",
        "# FROM THIS PICK A LEARNING RATE"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1e-08, 0.0001, 0.0, 30.0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "4uh-97bpLZCA",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "8d7c8880-5c9f-4623-f3ef-a758424451a3"
      },
      "source": [
        "tf.keras.backend.clear_session()\n",
        "tf.random.set_seed(51)\n",
        "np.random.seed(51)\n",
        "\n",
        "tf.keras.backend.clear_session()\n",
        "dataset = windowed_dataset(x_train, window_size, batch_size, shuffle_buffer_size)\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "  tf.keras.layers.Lambda(lambda x: tf.expand_dims(x,axis=-1),input_shape=[None]),\n",
        "  tf.keras.layers.SimpleRNN(40,return_sequences=True),\n",
        "  tf.keras.layers.SimpleRNN(40),\n",
        "  tf.keras.layers.Dense(1),\n",
        "  # YOUR CODE HERE\n",
        "  tf.keras.layers.Lambda(lambda x: x*100.0)\n",
        "])\n",
        "\n",
        "model.compile(loss=\"mse\", optimizer=tf.keras.optimizers.SGD(lr=1e-6, momentum=0.9),metrics=[\"mae\"])\n",
        "history = model.fit(dataset,epochs=500,verbose=1)\n",
        "                                                            \n",
        "# FIND A MODEL AND A LR THAT TRAINS TO AN MAE < 3                                                          "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/500\n",
            "94/94 [==============================] - 2s 24ms/step - loss: 1337.6217 - mae: 19.9555\n",
            "Epoch 2/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 57.0645 - mae: 5.2800\n",
            "Epoch 3/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 39.7193 - mae: 4.5086\n",
            "Epoch 4/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 29.7431 - mae: 3.7787\n",
            "Epoch 5/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 31.8450 - mae: 3.9747\n",
            "Epoch 6/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 31.2121 - mae: 3.9825\n",
            "Epoch 7/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 28.1187 - mae: 3.7656\n",
            "Epoch 8/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 34.2919 - mae: 4.2793\n",
            "Epoch 9/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 29.6721 - mae: 3.9083\n",
            "Epoch 10/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.9034 - mae: 3.8510\n",
            "Epoch 11/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 28.8343 - mae: 3.8042\n",
            "Epoch 12/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 30.4832 - mae: 4.0030\n",
            "Epoch 13/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.4489 - mae: 3.6144\n",
            "Epoch 14/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 31.0335 - mae: 4.0391\n",
            "Epoch 15/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 34.0601 - mae: 4.2817\n",
            "Epoch 16/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 27.2259 - mae: 3.7008\n",
            "Epoch 17/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 34.4859 - mae: 4.3172\n",
            "Epoch 18/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.9305 - mae: 3.5812\n",
            "Epoch 19/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 29.9314 - mae: 3.9368\n",
            "Epoch 20/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 31.4099 - mae: 4.0650\n",
            "Epoch 21/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 27.9719 - mae: 3.7519\n",
            "Epoch 22/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 27.2861 - mae: 3.6911\n",
            "Epoch 23/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 25.6785 - mae: 3.5038\n",
            "Epoch 24/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 31.2626 - mae: 4.0434\n",
            "Epoch 25/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.2963 - mae: 3.5166\n",
            "Epoch 26/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.6246 - mae: 3.6085\n",
            "Epoch 27/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.2828 - mae: 3.6153\n",
            "Epoch 28/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 28.4941 - mae: 3.7907\n",
            "Epoch 29/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 31.1021 - mae: 4.0177\n",
            "Epoch 30/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.0857 - mae: 3.7776\n",
            "Epoch 31/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 25.4108 - mae: 3.5225\n",
            "Epoch 32/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 26.2872 - mae: 3.5706\n",
            "Epoch 33/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.0750 - mae: 3.7944\n",
            "Epoch 34/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.8383 - mae: 3.6397\n",
            "Epoch 35/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.9690 - mae: 3.8737\n",
            "Epoch 36/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.2878 - mae: 3.5778\n",
            "Epoch 37/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 30.3952 - mae: 3.9153\n",
            "Epoch 38/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.7120 - mae: 3.4912\n",
            "Epoch 39/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.7384 - mae: 3.8213\n",
            "Epoch 40/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 33.7766 - mae: 4.2357\n",
            "Epoch 41/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 30.1029 - mae: 3.9341\n",
            "Epoch 42/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.0239 - mae: 3.4772\n",
            "Epoch 43/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 27.3786 - mae: 3.7229\n",
            "Epoch 44/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.3723 - mae: 3.4042\n",
            "Epoch 45/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 27.8596 - mae: 3.7494\n",
            "Epoch 46/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 27.3956 - mae: 3.6983\n",
            "Epoch 47/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.6032 - mae: 3.6675\n",
            "Epoch 48/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 36.4792 - mae: 4.4278\n",
            "Epoch 49/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.2820 - mae: 3.4260\n",
            "Epoch 50/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.3625 - mae: 3.5152\n",
            "Epoch 51/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.1871 - mae: 3.8186\n",
            "Epoch 52/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 25.2815 - mae: 3.4887\n",
            "Epoch 53/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.1534 - mae: 3.5920\n",
            "Epoch 54/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.3685 - mae: 3.6085\n",
            "Epoch 55/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 24.4912 - mae: 3.4342\n",
            "Epoch 56/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 26.3339 - mae: 3.5960\n",
            "Epoch 57/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.9236 - mae: 3.6068\n",
            "Epoch 58/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.3002 - mae: 3.4886\n",
            "Epoch 59/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.8121 - mae: 3.5624\n",
            "Epoch 60/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 27.5459 - mae: 3.7317\n",
            "Epoch 61/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 25.4922 - mae: 3.4980\n",
            "Epoch 62/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.3135 - mae: 3.5017\n",
            "Epoch 63/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.8599 - mae: 3.6402\n",
            "Epoch 64/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 28.9172 - mae: 3.8916\n",
            "Epoch 65/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.7388 - mae: 3.5741\n",
            "Epoch 66/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.9967 - mae: 3.5637\n",
            "Epoch 67/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.1627 - mae: 3.5738\n",
            "Epoch 68/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.0591 - mae: 3.3586\n",
            "Epoch 69/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5509 - mae: 3.3094\n",
            "Epoch 70/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.6895 - mae: 3.5337\n",
            "Epoch 71/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.1143 - mae: 3.3917\n",
            "Epoch 72/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 27.1058 - mae: 3.7175\n",
            "Epoch 73/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.8058 - mae: 3.6904\n",
            "Epoch 74/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6480 - mae: 3.3290\n",
            "Epoch 75/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6724 - mae: 3.3345\n",
            "Epoch 76/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.8125 - mae: 3.8413\n",
            "Epoch 77/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.6049 - mae: 3.5773\n",
            "Epoch 78/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.2786 - mae: 3.5308\n",
            "Epoch 79/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.4520 - mae: 3.4077\n",
            "Epoch 80/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.0305 - mae: 3.3643\n",
            "Epoch 81/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.6458 - mae: 3.4496\n",
            "Epoch 82/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.3795 - mae: 3.3177\n",
            "Epoch 83/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.6776 - mae: 3.5568\n",
            "Epoch 84/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.7020 - mae: 3.5573\n",
            "Epoch 85/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5805 - mae: 3.3062\n",
            "Epoch 86/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.2480 - mae: 3.3087\n",
            "Epoch 87/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.6576 - mae: 3.3215\n",
            "Epoch 88/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.8356 - mae: 3.3687\n",
            "Epoch 89/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.4028 - mae: 3.2177\n",
            "Epoch 90/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 25.3781 - mae: 3.4925\n",
            "Epoch 91/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.7962 - mae: 3.3389\n",
            "Epoch 92/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.4897 - mae: 3.3346\n",
            "Epoch 93/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.7510 - mae: 3.4426\n",
            "Epoch 94/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8472 - mae: 3.2589\n",
            "Epoch 95/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.9347 - mae: 3.3899\n",
            "Epoch 96/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.6331 - mae: 3.4639\n",
            "Epoch 97/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.7477 - mae: 3.4410\n",
            "Epoch 98/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.8600 - mae: 3.4837\n",
            "Epoch 99/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.3714 - mae: 3.5755\n",
            "Epoch 100/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.7819 - mae: 3.3836\n",
            "Epoch 101/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.2319 - mae: 3.6040\n",
            "Epoch 102/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5972 - mae: 3.3436\n",
            "Epoch 103/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.1093 - mae: 3.3671\n",
            "Epoch 104/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.2363 - mae: 3.4309\n",
            "Epoch 105/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.0927 - mae: 3.5929\n",
            "Epoch 106/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 41.9335 - mae: 4.6042\n",
            "Epoch 107/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.8929 - mae: 3.3813\n",
            "Epoch 108/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8680 - mae: 3.3885\n",
            "Epoch 109/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.4580 - mae: 3.3441\n",
            "Epoch 110/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.0511 - mae: 3.3787\n",
            "Epoch 111/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.5339 - mae: 3.4377\n",
            "Epoch 112/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.9606 - mae: 3.2871\n",
            "Epoch 113/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.5763 - mae: 3.2377\n",
            "Epoch 114/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.7542 - mae: 3.2675\n",
            "Epoch 115/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.4659 - mae: 3.2551\n",
            "Epoch 116/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.0524 - mae: 3.3640\n",
            "Epoch 117/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.5497 - mae: 3.4653\n",
            "Epoch 118/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.8076 - mae: 3.3797\n",
            "Epoch 119/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 27.2591 - mae: 3.7470\n",
            "Epoch 120/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.5903 - mae: 3.3599\n",
            "Epoch 121/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.7562 - mae: 3.4918\n",
            "Epoch 122/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.7994 - mae: 3.4954\n",
            "Epoch 123/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1650 - mae: 3.2997\n",
            "Epoch 124/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 27.6943 - mae: 3.7509\n",
            "Epoch 125/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.9625 - mae: 3.5407\n",
            "Epoch 126/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.0195 - mae: 3.4136\n",
            "Epoch 127/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.9546 - mae: 3.5980\n",
            "Epoch 128/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.5295 - mae: 3.5690\n",
            "Epoch 129/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.1720 - mae: 3.5131\n",
            "Epoch 130/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.0009 - mae: 3.7924\n",
            "Epoch 131/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.4961 - mae: 3.4279\n",
            "Epoch 132/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.9753 - mae: 3.4175\n",
            "Epoch 133/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5462 - mae: 3.3726\n",
            "Epoch 134/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1875 - mae: 3.3141\n",
            "Epoch 135/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.9177 - mae: 3.2474\n",
            "Epoch 136/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 34.8178 - mae: 4.2596\n",
            "Epoch 137/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 29.3823 - mae: 3.8675\n",
            "Epoch 138/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.2453 - mae: 3.6135\n",
            "Epoch 139/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.5994 - mae: 3.3495\n",
            "Epoch 140/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.5672 - mae: 3.1704\n",
            "Epoch 141/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 36.1057 - mae: 4.3963\n",
            "Epoch 142/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8146 - mae: 3.2353\n",
            "Epoch 143/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1073 - mae: 3.3077\n",
            "Epoch 144/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.1248 - mae: 3.4228\n",
            "Epoch 145/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.7128 - mae: 3.5542\n",
            "Epoch 146/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.2509 - mae: 3.3424\n",
            "Epoch 147/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.8679 - mae: 3.5912\n",
            "Epoch 148/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5227 - mae: 3.3554\n",
            "Epoch 149/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1256 - mae: 3.3800\n",
            "Epoch 150/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 30.0633 - mae: 3.9264\n",
            "Epoch 151/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.5604 - mae: 3.4925\n",
            "Epoch 152/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.7617 - mae: 3.3776\n",
            "Epoch 153/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.5258 - mae: 3.4750\n",
            "Epoch 154/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.4052 - mae: 3.2291\n",
            "Epoch 155/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.2623 - mae: 3.5497\n",
            "Epoch 156/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.3243 - mae: 3.2271\n",
            "Epoch 157/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.2897 - mae: 3.6187\n",
            "Epoch 158/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6706 - mae: 3.3718\n",
            "Epoch 159/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.3719 - mae: 3.3063\n",
            "Epoch 160/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.4274 - mae: 3.3612\n",
            "Epoch 161/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.9259 - mae: 3.2692\n",
            "Epoch 162/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.1161 - mae: 3.3846\n",
            "Epoch 163/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.9432 - mae: 3.5228\n",
            "Epoch 164/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.4975 - mae: 3.3495\n",
            "Epoch 165/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8090 - mae: 3.3112\n",
            "Epoch 166/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.5537 - mae: 3.6342\n",
            "Epoch 167/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6836 - mae: 3.3811\n",
            "Epoch 168/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.8141 - mae: 3.4160\n",
            "Epoch 169/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.7179 - mae: 3.4057\n",
            "Epoch 170/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.8873 - mae: 3.2922\n",
            "Epoch 171/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.3015 - mae: 3.1131\n",
            "Epoch 172/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.2313 - mae: 3.3247\n",
            "Epoch 173/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.7439 - mae: 3.4897\n",
            "Epoch 174/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.3637 - mae: 3.3023\n",
            "Epoch 175/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8573 - mae: 3.4010\n",
            "Epoch 176/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.4133 - mae: 3.3471\n",
            "Epoch 177/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5714 - mae: 3.3524\n",
            "Epoch 178/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5172 - mae: 3.2455\n",
            "Epoch 179/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.3268 - mae: 3.6429\n",
            "Epoch 180/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.5660 - mae: 3.6847\n",
            "Epoch 181/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.9850 - mae: 3.2720\n",
            "Epoch 182/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 28.3189 - mae: 3.8432\n",
            "Epoch 183/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.7872 - mae: 3.3013\n",
            "Epoch 184/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.7209 - mae: 3.3895\n",
            "Epoch 185/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5398 - mae: 3.3367\n",
            "Epoch 186/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.9812 - mae: 3.3770\n",
            "Epoch 187/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.8189 - mae: 3.4723\n",
            "Epoch 188/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.2888 - mae: 3.2311\n",
            "Epoch 189/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.9970 - mae: 3.4314\n",
            "Epoch 190/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.0609 - mae: 3.5981\n",
            "Epoch 191/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.6814 - mae: 3.2874\n",
            "Epoch 192/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.9366 - mae: 3.4087\n",
            "Epoch 193/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.9633 - mae: 3.4041\n",
            "Epoch 194/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.5210 - mae: 3.3526\n",
            "Epoch 195/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.1751 - mae: 3.4470\n",
            "Epoch 196/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.4232 - mae: 3.3393\n",
            "Epoch 197/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 28.4424 - mae: 3.8623\n",
            "Epoch 198/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.4579 - mae: 3.2603\n",
            "Epoch 199/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.1384 - mae: 3.2745\n",
            "Epoch 200/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.9736 - mae: 3.2977\n",
            "Epoch 201/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.1058 - mae: 3.4397\n",
            "Epoch 202/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.0412 - mae: 3.4362\n",
            "Epoch 203/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.2812 - mae: 3.3483\n",
            "Epoch 204/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.2431 - mae: 3.4189\n",
            "Epoch 205/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.8962 - mae: 3.5050\n",
            "Epoch 206/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 22.1117 - mae: 3.2394\n",
            "Epoch 207/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 24.1636 - mae: 3.4066\n",
            "Epoch 208/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.4698 - mae: 3.4013\n",
            "Epoch 209/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.7828 - mae: 3.7129\n",
            "Epoch 210/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.3043 - mae: 3.2467\n",
            "Epoch 211/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.9562 - mae: 3.2130\n",
            "Epoch 212/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.3539 - mae: 3.3370\n",
            "Epoch 213/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.6876 - mae: 3.2733\n",
            "Epoch 214/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5851 - mae: 3.2577\n",
            "Epoch 215/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1579 - mae: 3.2914\n",
            "Epoch 216/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 28.7103 - mae: 3.8783\n",
            "Epoch 217/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 35.0029 - mae: 4.2688\n",
            "Epoch 218/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.7697 - mae: 3.3820\n",
            "Epoch 219/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.3078 - mae: 3.5384\n",
            "Epoch 220/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1622 - mae: 3.3622\n",
            "Epoch 221/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.5606 - mae: 3.4817\n",
            "Epoch 222/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 22.1108 - mae: 3.2194\n",
            "Epoch 223/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.4420 - mae: 3.3122\n",
            "Epoch 224/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.7482 - mae: 3.1896\n",
            "Epoch 225/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 21.6131 - mae: 3.1706\n",
            "Epoch 226/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.0404 - mae: 3.4128\n",
            "Epoch 227/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.0124 - mae: 3.3548\n",
            "Epoch 228/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.3447 - mae: 3.3342\n",
            "Epoch 229/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.9336 - mae: 3.2891\n",
            "Epoch 230/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.8925 - mae: 3.4722\n",
            "Epoch 231/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.0201 - mae: 3.3418\n",
            "Epoch 232/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.3613 - mae: 3.4511\n",
            "Epoch 233/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 28.5728 - mae: 3.8590\n",
            "Epoch 234/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.1373 - mae: 3.4872\n",
            "Epoch 235/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 29.0654 - mae: 3.8340\n",
            "Epoch 236/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.2828 - mae: 3.3352\n",
            "Epoch 237/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.2052 - mae: 3.5144\n",
            "Epoch 238/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.1445 - mae: 3.2401\n",
            "Epoch 239/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.8469 - mae: 3.5613\n",
            "Epoch 240/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.4937 - mae: 3.1478\n",
            "Epoch 241/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1144 - mae: 3.3500\n",
            "Epoch 242/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.3822 - mae: 3.2662\n",
            "Epoch 243/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1901 - mae: 3.3416\n",
            "Epoch 244/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5006 - mae: 3.3878\n",
            "Epoch 245/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.3101 - mae: 3.2649\n",
            "Epoch 246/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.0038 - mae: 3.3790\n",
            "Epoch 247/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.6974 - mae: 3.5384\n",
            "Epoch 248/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.4125 - mae: 3.4419\n",
            "Epoch 249/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.7955 - mae: 3.5101\n",
            "Epoch 250/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.9233 - mae: 3.4985\n",
            "Epoch 251/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8604 - mae: 3.3145\n",
            "Epoch 252/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 28.0872 - mae: 3.8206\n",
            "Epoch 253/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1195 - mae: 3.2927\n",
            "Epoch 254/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.0258 - mae: 3.6038\n",
            "Epoch 255/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5479 - mae: 3.3891\n",
            "Epoch 256/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.2000 - mae: 3.2523\n",
            "Epoch 257/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.4034 - mae: 3.3736\n",
            "Epoch 258/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6011 - mae: 3.4064\n",
            "Epoch 259/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.4074 - mae: 3.3304\n",
            "Epoch 260/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.2903 - mae: 3.5262\n",
            "Epoch 261/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.6461 - mae: 3.2809\n",
            "Epoch 262/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.0249 - mae: 3.2197\n",
            "Epoch 263/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.8809 - mae: 3.1978\n",
            "Epoch 264/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8196 - mae: 3.3077\n",
            "Epoch 265/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.7799 - mae: 3.2150\n",
            "Epoch 266/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.4955 - mae: 3.3366\n",
            "Epoch 267/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.7716 - mae: 3.3757\n",
            "Epoch 268/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.3238 - mae: 3.2658\n",
            "Epoch 269/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1091 - mae: 3.2857\n",
            "Epoch 270/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.9885 - mae: 3.4101\n",
            "Epoch 271/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.0117 - mae: 3.6411\n",
            "Epoch 272/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.2901 - mae: 3.4551\n",
            "Epoch 273/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.9590 - mae: 3.3224\n",
            "Epoch 274/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.2866 - mae: 3.5742\n",
            "Epoch 275/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.4581 - mae: 3.2528\n",
            "Epoch 276/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.6189 - mae: 3.1824\n",
            "Epoch 277/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.8638 - mae: 3.5681\n",
            "Epoch 278/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.4391 - mae: 3.2283\n",
            "Epoch 279/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.9884 - mae: 3.2995\n",
            "Epoch 280/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 27.3423 - mae: 3.7610\n",
            "Epoch 281/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5672 - mae: 3.2464\n",
            "Epoch 282/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.2447 - mae: 3.4190\n",
            "Epoch 283/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.3331 - mae: 3.4308\n",
            "Epoch 284/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6025 - mae: 3.3413\n",
            "Epoch 285/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5260 - mae: 3.2413\n",
            "Epoch 286/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6670 - mae: 3.3689\n",
            "Epoch 287/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.2390 - mae: 3.2132\n",
            "Epoch 288/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 28.0824 - mae: 3.8141\n",
            "Epoch 289/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.4614 - mae: 3.4470\n",
            "Epoch 290/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.0380 - mae: 3.4212\n",
            "Epoch 291/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.9112 - mae: 3.4698\n",
            "Epoch 292/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 25.2624 - mae: 3.5562\n",
            "Epoch 293/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.4076 - mae: 3.4365\n",
            "Epoch 294/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.5131 - mae: 3.4615\n",
            "Epoch 295/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.5469 - mae: 3.2535\n",
            "Epoch 296/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.6684 - mae: 3.2754\n",
            "Epoch 297/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.6885 - mae: 3.2144\n",
            "Epoch 298/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1028 - mae: 3.3158\n",
            "Epoch 299/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.5370 - mae: 3.3876\n",
            "Epoch 300/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 22.3225 - mae: 3.2408\n",
            "Epoch 301/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.7127 - mae: 3.5404\n",
            "Epoch 302/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.9393 - mae: 3.4265\n",
            "Epoch 303/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.8012 - mae: 3.4012\n",
            "Epoch 304/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.8210 - mae: 3.1754\n",
            "Epoch 305/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.6642 - mae: 3.5573\n",
            "Epoch 306/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.2852 - mae: 3.2613\n",
            "Epoch 307/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5959 - mae: 3.3197\n",
            "Epoch 308/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.1913 - mae: 3.3919\n",
            "Epoch 309/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.7100 - mae: 3.1553\n",
            "Epoch 310/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.3694 - mae: 3.1636\n",
            "Epoch 311/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.6478 - mae: 3.2915\n",
            "Epoch 312/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 20.8870 - mae: 3.1616\n",
            "Epoch 313/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 29.5674 - mae: 3.9464\n",
            "Epoch 314/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5914 - mae: 3.3978\n",
            "Epoch 315/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.0853 - mae: 3.5847\n",
            "Epoch 316/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.9348 - mae: 3.5001\n",
            "Epoch 317/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.6693 - mae: 3.5016\n",
            "Epoch 318/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.7155 - mae: 3.2937\n",
            "Epoch 319/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.9949 - mae: 3.1897\n",
            "Epoch 320/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.7782 - mae: 3.2958\n",
            "Epoch 321/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.4441 - mae: 3.4724\n",
            "Epoch 322/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.3152 - mae: 3.2562\n",
            "Epoch 323/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8794 - mae: 3.3059\n",
            "Epoch 324/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 20.9075 - mae: 3.1715\n",
            "Epoch 325/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 30.2827 - mae: 3.9945\n",
            "Epoch 326/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.2000 - mae: 3.3972\n",
            "Epoch 327/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.2228 - mae: 3.3546\n",
            "Epoch 328/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.5232 - mae: 3.3875\n",
            "Epoch 329/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.3772 - mae: 3.5542\n",
            "Epoch 330/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.3731 - mae: 3.3516\n",
            "Epoch 331/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.5491 - mae: 3.3646\n",
            "Epoch 332/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.0456 - mae: 3.2385\n",
            "Epoch 333/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 30.0783 - mae: 3.9632\n",
            "Epoch 334/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.2826 - mae: 3.3322\n",
            "Epoch 335/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.4398 - mae: 3.2593\n",
            "Epoch 336/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.9195 - mae: 3.4085\n",
            "Epoch 337/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.0232 - mae: 3.6117\n",
            "Epoch 338/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.1997 - mae: 3.3918\n",
            "Epoch 339/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.0349 - mae: 3.4941\n",
            "Epoch 340/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.7435 - mae: 3.4019\n",
            "Epoch 341/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.6412 - mae: 3.6238\n",
            "Epoch 342/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1630 - mae: 3.3043\n",
            "Epoch 343/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.0925 - mae: 3.2670\n",
            "Epoch 344/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5505 - mae: 3.3851\n",
            "Epoch 345/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.9367 - mae: 3.2288\n",
            "Epoch 346/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5957 - mae: 3.4011\n",
            "Epoch 347/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.8170 - mae: 3.6196\n",
            "Epoch 348/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 29.3182 - mae: 3.9502\n",
            "Epoch 349/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.9623 - mae: 3.6026\n",
            "Epoch 350/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.8611 - mae: 3.2006\n",
            "Epoch 351/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5008 - mae: 3.2531\n",
            "Epoch 352/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1704 - mae: 3.3754\n",
            "Epoch 353/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.6558 - mae: 3.5527\n",
            "Epoch 354/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.4556 - mae: 3.2609\n",
            "Epoch 355/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.8850 - mae: 3.3825\n",
            "Epoch 356/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.6756 - mae: 3.3758\n",
            "Epoch 357/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5272 - mae: 3.3797\n",
            "Epoch 358/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 29.7278 - mae: 3.8734\n",
            "Epoch 359/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 22.6378 - mae: 3.2854\n",
            "Epoch 360/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.7193 - mae: 3.4995\n",
            "Epoch 361/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.4980 - mae: 3.1370\n",
            "Epoch 362/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.6238 - mae: 3.4856\n",
            "Epoch 363/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.5323 - mae: 3.2899\n",
            "Epoch 364/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.0848 - mae: 3.2984\n",
            "Epoch 365/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.6173 - mae: 3.2610\n",
            "Epoch 366/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.7690 - mae: 3.4191\n",
            "Epoch 367/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.6145 - mae: 3.5050\n",
            "Epoch 368/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.2665 - mae: 3.2322\n",
            "Epoch 369/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.5553 - mae: 3.3969\n",
            "Epoch 370/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.1065 - mae: 3.1266\n",
            "Epoch 371/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1463 - mae: 3.3110\n",
            "Epoch 372/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8000 - mae: 3.3931\n",
            "Epoch 373/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.4148 - mae: 3.1449\n",
            "Epoch 374/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.0603 - mae: 3.4478\n",
            "Epoch 375/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.1301 - mae: 3.4326\n",
            "Epoch 376/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.9409 - mae: 3.4118\n",
            "Epoch 377/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1628 - mae: 3.3278\n",
            "Epoch 378/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.4283 - mae: 3.3640\n",
            "Epoch 379/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.7222 - mae: 3.1802\n",
            "Epoch 380/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.4135 - mae: 3.4067\n",
            "Epoch 381/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.5052 - mae: 3.5136\n",
            "Epoch 382/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 21.5175 - mae: 3.1727\n",
            "Epoch 383/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.3678 - mae: 3.1602\n",
            "Epoch 384/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.8389 - mae: 3.5600\n",
            "Epoch 385/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.4898 - mae: 3.1507\n",
            "Epoch 386/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.5742 - mae: 3.2462\n",
            "Epoch 387/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.1906 - mae: 3.2577\n",
            "Epoch 388/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.1823 - mae: 3.1297\n",
            "Epoch 389/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.3700 - mae: 3.2641\n",
            "Epoch 390/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 24.8839 - mae: 3.4752\n",
            "Epoch 391/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.9115 - mae: 3.2177\n",
            "Epoch 392/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.0204 - mae: 3.4146\n",
            "Epoch 393/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.7763 - mae: 3.3087\n",
            "Epoch 394/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.7429 - mae: 3.1971\n",
            "Epoch 395/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.0399 - mae: 3.4636\n",
            "Epoch 396/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.6505 - mae: 3.3001\n",
            "Epoch 397/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.1312 - mae: 3.4167\n",
            "Epoch 398/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8077 - mae: 3.4468\n",
            "Epoch 399/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 20.8474 - mae: 3.1689\n",
            "Epoch 400/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 30.7198 - mae: 3.9030\n",
            "Epoch 401/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8424 - mae: 3.3637\n",
            "Epoch 402/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 26.4477 - mae: 3.6435\n",
            "Epoch 403/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.8193 - mae: 3.1759\n",
            "Epoch 404/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.2927 - mae: 3.1387\n",
            "Epoch 405/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1989 - mae: 3.3601\n",
            "Epoch 406/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.0374 - mae: 3.3384\n",
            "Epoch 407/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.4797 - mae: 3.2407\n",
            "Epoch 408/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.8385 - mae: 3.2032\n",
            "Epoch 409/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.1950 - mae: 3.1469\n",
            "Epoch 410/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8303 - mae: 3.4180\n",
            "Epoch 411/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.0788 - mae: 3.2322\n",
            "Epoch 412/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.2700 - mae: 3.3382\n",
            "Epoch 413/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.3401 - mae: 3.3823\n",
            "Epoch 414/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.1031 - mae: 3.2584\n",
            "Epoch 415/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.7491 - mae: 3.2918\n",
            "Epoch 416/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.8766 - mae: 3.1921\n",
            "Epoch 417/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.3757 - mae: 3.3433\n",
            "Epoch 418/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.4814 - mae: 3.5675\n",
            "Epoch 419/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.3541 - mae: 3.2299\n",
            "Epoch 420/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.9468 - mae: 3.3367\n",
            "Epoch 421/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.8647 - mae: 3.2280\n",
            "Epoch 422/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.1712 - mae: 3.2542\n",
            "Epoch 423/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.8806 - mae: 3.3017\n",
            "Epoch 424/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.1943 - mae: 3.6786\n",
            "Epoch 425/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.4469 - mae: 3.4640\n",
            "Epoch 426/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1646 - mae: 3.3586\n",
            "Epoch 427/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5580 - mae: 3.2743\n",
            "Epoch 428/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.3654 - mae: 3.3950\n",
            "Epoch 429/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.7642 - mae: 3.2098\n",
            "Epoch 430/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.1690 - mae: 3.2311\n",
            "Epoch 431/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8304 - mae: 3.2923\n",
            "Epoch 432/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 20.9426 - mae: 3.0885\n",
            "Epoch 433/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.4779 - mae: 3.1834\n",
            "Epoch 434/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8252 - mae: 3.4145\n",
            "Epoch 435/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.5405 - mae: 3.4017\n",
            "Epoch 436/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.1582 - mae: 3.3714\n",
            "Epoch 437/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.9940 - mae: 3.6040\n",
            "Epoch 438/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.1266 - mae: 3.1446\n",
            "Epoch 439/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.0879 - mae: 3.4969\n",
            "Epoch 440/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.8170 - mae: 3.2057\n",
            "Epoch 441/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.1567 - mae: 3.2266\n",
            "Epoch 442/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.2356 - mae: 3.3435\n",
            "Epoch 443/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.5278 - mae: 3.2645\n",
            "Epoch 444/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.2448 - mae: 3.1546\n",
            "Epoch 445/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.0379 - mae: 3.4150\n",
            "Epoch 446/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.4779 - mae: 3.1527\n",
            "Epoch 447/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.8913 - mae: 3.4177\n",
            "Epoch 448/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 23.3124 - mae: 3.3331\n",
            "Epoch 449/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.8988 - mae: 3.1940\n",
            "Epoch 450/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.0312 - mae: 3.2678\n",
            "Epoch 451/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.8575 - mae: 3.3117\n",
            "Epoch 452/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.9065 - mae: 3.4026\n",
            "Epoch 453/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.7757 - mae: 3.1759\n",
            "Epoch 454/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.1467 - mae: 3.2265\n",
            "Epoch 455/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.7878 - mae: 3.2172\n",
            "Epoch 456/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.7424 - mae: 3.4694\n",
            "Epoch 457/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.5433 - mae: 3.4615\n",
            "Epoch 458/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 21.3384 - mae: 3.1311\n",
            "Epoch 459/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.9736 - mae: 3.3465\n",
            "Epoch 460/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.7135 - mae: 3.2074\n",
            "Epoch 461/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.4582 - mae: 3.2536\n",
            "Epoch 462/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.7782 - mae: 3.2915\n",
            "Epoch 463/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.2848 - mae: 3.3668\n",
            "Epoch 464/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 25.8990 - mae: 3.6086\n",
            "Epoch 465/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 25.0948 - mae: 3.5544\n",
            "Epoch 466/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.9112 - mae: 3.2315\n",
            "Epoch 467/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.6743 - mae: 3.2044\n",
            "Epoch 468/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.3859 - mae: 3.2608\n",
            "Epoch 469/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.2646 - mae: 3.2279\n",
            "Epoch 470/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.8813 - mae: 3.2836\n",
            "Epoch 471/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.1467 - mae: 3.4481\n",
            "Epoch 472/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.6833 - mae: 3.3726\n",
            "Epoch 473/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 20.9061 - mae: 3.1077\n",
            "Epoch 474/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.7551 - mae: 3.4216\n",
            "Epoch 475/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.2953 - mae: 3.3470\n",
            "Epoch 476/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.5735 - mae: 3.1571\n",
            "Epoch 477/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.5966 - mae: 3.2538\n",
            "Epoch 478/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.4688 - mae: 3.6851\n",
            "Epoch 479/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.8404 - mae: 3.1903\n",
            "Epoch 480/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 25.0797 - mae: 3.4948\n",
            "Epoch 481/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 23.1062 - mae: 3.3113\n",
            "Epoch 482/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.0811 - mae: 3.1368\n",
            "Epoch 483/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.3996 - mae: 3.1527\n",
            "Epoch 484/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.5417 - mae: 3.5032\n",
            "Epoch 485/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.6258 - mae: 3.2792\n",
            "Epoch 486/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 21.1794 - mae: 3.1420\n",
            "Epoch 487/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 24.2316 - mae: 3.4644\n",
            "Epoch 488/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 20.9997 - mae: 3.1114\n",
            "Epoch 489/500\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 22.1692 - mae: 3.2474\n",
            "Epoch 490/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.6902 - mae: 3.5228\n",
            "Epoch 491/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 22.0289 - mae: 3.2302\n",
            "Epoch 492/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.4574 - mae: 3.4520\n",
            "Epoch 493/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 24.0629 - mae: 3.3970\n",
            "Epoch 494/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 23.2011 - mae: 3.3752\n",
            "Epoch 495/500\n",
            "94/94 [==============================] - 1s 9ms/step - loss: 24.8225 - mae: 3.5356\n",
            "Epoch 496/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 21.8473 - mae: 3.1864\n",
            "Epoch 497/500\n",
            "94/94 [==============================] - 1s 11ms/step - loss: 21.3352 - mae: 3.1605\n",
            "Epoch 498/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 21.5449 - mae: 3.1957\n",
            "Epoch 499/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 22.1431 - mae: 3.2632\n",
            "Epoch 500/500\n",
            "94/94 [==============================] - 1s 10ms/step - loss: 26.9976 - mae: 3.7140\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "icGDaND7z0ne",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 388
        },
        "outputId": "e611805f-d627-4b57-a5be-3bf438112aa9"
      },
      "source": [
        "forecast = []\n",
        "results = []\n",
        "for time in range(len(series) - window_size):\n",
        "  forecast.append(model.predict(series[time:time + window_size][np.newaxis]))\n",
        "\n",
        "forecast = forecast[split_time-window_size:]\n",
        "results = np.array(forecast)[:, 0, 0]\n",
        "\n",
        "\n",
        "plt.figure(figsize=(10, 6))\n",
        "\n",
        "plot_series(time_valid, x_valid)\n",
        "plot_series(time_valid, results)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "KfPeqI7rz4LD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "19804a77-4b03-4499-a408-2b97df01b07f"
      },
      "source": [
        "tf.keras.metrics.mean_absolute_error(x_valid, results).numpy()\n",
        "\n",
        "# YOUR RESULT HERE SHOULD BE LESS THAN 4"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "3.4177802"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "JUsdZB_tzDLe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "outputId": "ceccee63-71bc-4aae-9ec9-832035fca7b1"
      },
      "source": [
        "import matplotlib.image  as mpimg\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "#-----------------------------------------------------------\n",
        "# Retrieve a list of list results on training and test data\n",
        "# sets for each training epoch\n",
        "#-----------------------------------------------------------\n",
        "mae=history.history['mae']\n",
        "loss=history.history['loss']\n",
        "\n",
        "epochs=range(len(loss)) # Get number of epochs\n",
        "\n",
        "#------------------------------------------------\n",
        "# Plot MAE and Loss\n",
        "#------------------------------------------------\n",
        "plt.plot(epochs, mae, 'r')\n",
        "plt.plot(epochs, loss, 'b')\n",
        "plt.title('MAE and Loss')\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.legend([\"MAE\", \"Loss\"])\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "epochs_zoom = epochs[200:]\n",
        "mae_zoom = mae[200:]\n",
        "loss_zoom = loss[200:]\n",
        "\n",
        "#------------------------------------------------\n",
        "# Plot Zoomed MAE and Loss\n",
        "#------------------------------------------------\n",
        "plt.plot(epochs_zoom, mae_zoom, 'r')\n",
        "plt.plot(epochs_zoom, loss_zoom, 'b')\n",
        "plt.title('MAE and Loss')\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.legend([\"MAE\", \"Loss\"])\n",
        "\n",
        "plt.figure()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ow_m30dYP8B0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}