LSTM_CRDF_S_(1).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.4"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "m9-24HcAn1tw"
      },
      "source": [
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras import layers\n",
        "import tensorflow.keras.layers as KL\n",
        "from PIL import Image\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "swfv4BbVErGh"
      },
      "source": [
        "\n",
        "# from google.colab import files\n",
        "# uploaded = files.upload()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BDWF4Q_jLL3I",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e9cc7fd6-b712-422b-f5e9-e940b851d4d7"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "00e8-aNpMEN7"
      },
      "source": [
        "# train=['nonfall_skel_1_RGB.mp4','nonfall_skel_2_RGB.mp4','fall_skel_1_RGB.mp4','fall_skel_2_RGB.mp4']\n",
        "# test=['nonfall_skel_3_RGB.mp4','fall_skel_3_RGB.mp4']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vQ85Z5NCEyma"
      },
      "source": [
        "\n",
        "import cv2\n",
        "location_m = '/content/drive/My Drive/Colab Notebooks/'\n",
        "location_s = '/content/drive/My Drive/'\n",
        "def data_collector(name,i,j):\n",
        "  images = []\n",
        "  target = []\n",
        "  count = 0\n",
        "  co = 0\n",
        "  for ele in range(i,j+1):\n",
        "    cap = cv2.VideoCapture(location_m+'Videos/'+name+'/'+'skel_'+str(ele)+'_RGB.avi')\n",
        "    frames = []\n",
        "    temp_target = []\n",
        "    while cap.isOpened():\n",
        "        ret, image = cap.read()\n",
        "        if not ret:\n",
        "          break\n",
        "        gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)\n",
        "        # cv2_imshow(image) # Note cv2_imshow, not cv2.imshow\n",
        "        frames.append(gray)\n",
        "        count = count + 1\n",
        "        if(name == 'Fall'):\n",
        "          temp_target.append(1)\n",
        "        else:\n",
        "          temp_target.append(0)\n",
        "    frames = np.array(frames)\n",
        "    temp_target = np.array(temp_target)\n",
        "    if(len(frames)>125):\n",
        "        frames = frames[-125:]\n",
        "        temp_target = temp_target[-125:]\n",
        "\n",
        "    images.append(frames)\n",
        "    target.append(temp_target)\n",
        "    co += 1\n",
        "    cv2.waitKey(1) & 0xff\n",
        "    cv2.destroyAllWindows()\n",
        "    cap.release()\n",
        "  images = np.array(images)\n",
        "\n",
        "  target = np.array(target)\n",
        "  return images , target"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P2Ryh4JAafEw"
      },
      "source": [
        "images_fall, target_fall = data_collector('Fall',1,30)\n",
        "images_not_fall , target_not_fall = data_collector('NotFall',1,13)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hmEEU4lSb9cZ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "2ad97fe4-5a6f-4c8d-dfb6-1ce2bda778d4"
      },
      "source": [
        "# np.squeeze(images_fall,axis = (2,))\n",
        "len(target_fall)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "30"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vq9JEa2R-6Kt",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "91b71942-ae91-4b29-ec8f-b4eec0774a4a"
      },
      "source": [
        "images_fall.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(30,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2b0Vxbdd9NLj"
      },
      "source": [
        "# print(count)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vW10Q_PdOs1h",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "1cad60c7-fd88-42ec-b2d7-2fca8c2a8518"
      },
      "source": [
        "images_fall_f = []\n",
        "for l in images_fall:\n",
        "    for f in l:\n",
        "        images_fall_f.append(f)\n",
        "\n",
        "images_fall_f = np.array(images_fall_f)\n",
        "print(images_fall_f.shape)\n",
        "target_fall_f = []\n",
        "for l in target_fall:\n",
        "    for f in l:\n",
        "        target_fall_f.append(f)\n",
        "\n",
        "target_fall_f = np.array(target_fall_f)\n",
        "print(target_fall_f.shape)\n",
        "\n",
        "images_not_fall_f = []\n",
        "for l in images_not_fall:\n",
        "    for f in l:\n",
        "        images_not_fall_f.append(f)\n",
        "\n",
        "images_not_fall_f = np.array(images_not_fall_f)\n",
        "print(images_not_fall_f.shape)\n",
        "target_not_fall_f = []\n",
        "for l in target_not_fall:\n",
        "    for f in l:\n",
        "        target_not_fall_f.append(f)\n",
        "\n",
        "target_not_fall_f = np.array(target_not_fall_f)\n",
        "print(target_not_fall_f.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(2618, 288, 384)\n",
            "(2618,)\n",
            "(1625, 288, 384)\n",
            "(1625,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UrtdSEI_6CRR"
      },
      "source": [
        "images_fall = images_fall_f\n",
        "target_fall = target_fall_f\n",
        "images_not_fall = images_not_fall_f\n",
        "target_not_fall = target_not_fall_f"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DSJOQMdYFtZA",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "3149c82f-9cfb-4a94-fcea-ae32d56dc216"
      },
      "source": [
        "\n",
        "from matplotlib import pyplot as plt\n",
        "plt.imshow(images_fall[134], interpolation='nearest')\n",
        "plt.show()"
      ],
      "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": {
        "id": "zSrYGfJ6eMCr"
      },
      "source": [
        "x = images_fall.shape[0] - int(images_fall.shape[0]*0.3)\n",
        "x_trainf = images_fall[:x]\n",
        "x_testf = images_fall[x:]\n",
        "y_trainf = target_fall[:x]\n",
        "y_testf = target_fall[x:]\n",
        "x = images_not_fall.shape[0] - int(images_not_fall.shape[0]*0.3)\n",
        "x_trainfn = images_not_fall[:x]\n",
        "x_testfn = images_not_fall[x:]\n",
        "y_trainfn = target_not_fall[:x]\n",
        "y_testfn = target_not_fall[x:]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dUruN6OzevEC",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        },
        "outputId": "03d0fea1-63eb-4734-c21d-6f464bda9297"
      },
      "source": [
        "print(x_trainf.shape)\n",
        "print(y_trainf.shape)\n",
        "print(x_trainfn.shape)\n",
        "print(y_trainfn.shape)\n",
        "print(x_testf.shape)\n",
        "print(y_testf.shape)\n",
        "print(x_testfn.shape)\n",
        "print(y_testfn.shape)\n",
        "x_train = np.append(x_trainf,x_trainfn,axis = 0)\n",
        "y_train = np.append(y_trainf,y_trainfn,axis = 0)\n",
        "x_test = np.append(x_testf,x_testfn,axis = 0)\n",
        "y_test = np.append(y_testf,y_testfn,axis = 0)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(1833, 288, 384)\n",
            "(1833,)\n",
            "(1138, 288, 384)\n",
            "(1138,)\n",
            "(785, 288, 384)\n",
            "(785,)\n",
            "(487, 288, 384)\n",
            "(487,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ewVGcYKU6CRm"
      },
      "source": [
        "x = x_test.shape[0] - int(x_test.shape[0]*0.5)\n",
        "x_val = x_test[:x]\n",
        "x_test = x_test[x:]\n",
        "y_val = y_test[:x]\n",
        "y_test = y_test[x:]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HdKDAnUrg2Tx",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        },
        "outputId": "9efbad01-8284-4eb9-cbb8-7de4fbb2a6cf"
      },
      "source": [
        "print(x_train.shape)\n",
        "print(y_train.shape)\n",
        "print(x_val.shape)\n",
        "print(y_val.shape)\n",
        "print(x_test.shape)\n",
        "print(y_test.shape)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(2971, 288, 384)\n",
            "(2971,)\n",
            "(636, 288, 384)\n",
            "(636,)\n",
            "(636, 288, 384)\n",
            "(636,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IEtP3kSEF-6F"
      },
      "source": [
        "# inputs = KL.Input(shape=(288, 384))\n",
        "# # For RNN\n",
        "# # x = KL.RNN(64, activation ='relu')(inputs)\n",
        "\n",
        "# # For LSTM\n",
        "# x = KL.LSTM(64)(inputs)\n",
        "# # x = KL.CuDNNGRU(64)(inputs)\n",
        "# # x=KL.Input(shape=(28,28))\n",
        "# outputs = KL.Dense(10, activation=\"softmax\")(x)\n",
        "\n",
        "# model = tf.keras.models.Model(inputs, outputs)\n",
        "# model.summary()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BO5FyfSA6CR2"
      },
      "source": [
        "# import tensorflow as tf\n",
        "# from tensorflow.keras.models import Sequential\n",
        "# from tensorflow.keras.layers import Dense,Flatten, Dropout\n",
        "# import tensorflow as tf\n",
        "# from tensorflow.keras.layers import Input, LSTM, ConvLSTM2D, GlobalMaxPool2D, concatenate, Dense, Dropout, Conv2D, Permute, MaxPooling2D, maximum, LeakyReLU, Reshape\n",
        "# from tensorflow.keras.models import Model\n",
        "# from tensorflow.keras import optimizers\n",
        "# from tensorflow import keras\n",
        "# from tensorflow.keras import backend as K\n",
        "# seq_len = 30\n",
        "# img_height = 288\n",
        "# img_width = 384\n",
        "# model = Sequential()\n",
        "# model.add(ConvLSTM2D(filters = 64, kernel_size = (3, 3), return_sequences = False, data_format = \"channels_last\", input_shape = (seq_len, img_height, img_width, 1)))\n",
        "# model.add(ConvLSTM2D(filters = 32, kernel_size = (3, 3), return_sequences = False))\n",
        "# model.add(Dropout(0.2))\n",
        "# model.add(Flatten())\n",
        "# model.add(Dense(256, activation=\"relu\"))\n",
        "# model.add(Dense(1, activation = \"sigmoid\"))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vO-Xrw9fGilH"
      },
      "source": [
        "# def modellstm():\n",
        "#   inp = KL.Input(shape=(288, 384))\n",
        "#   regress = KL.LSTM(512, return_sequences=True)(inp)\n",
        "#   regress = KL.LSTM(128, return_sequences=True)(regress)\n",
        "#   regress = KL.LSTM(64, return_sequences=False)(regress)\n",
        "#   regress = KL.Dense(4, activation='relu')(regress)\n",
        "\n",
        "#   return tf.keras.models.Model(inp, regress)\n",
        "# m=modellstm()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_UdNoR7QHcIL"
      },
      "source": [
        "def modellstm():\n",
        "  inp = KL.Input(shape=(288, 384))\n",
        "  regress = KL.LSTM(512, return_sequences=True)(inp)\n",
        "  regress = KL.LSTM(678, return_sequences=True)(regress)\n",
        "  regress = KL.LSTM(430, return_sequences=True)(regress)\n",
        "  regress = KL.LSTM(130, return_sequences=False)(regress)\n",
        "  regress = KL.Dense(1, activation='sigmoid')(regress)\n",
        "  return tf.keras.models.Model(inp, regress)\n",
        "m=modellstm()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "irqXkKeM6CSD"
      },
      "source": [
        "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\n",
        "keras_callbacks = [ EarlyStopping(\n",
        "   monitor='val_loss',\n",
        "   patience=5,\n",
        "   min_delta=0.001,\n",
        "   mode='auto',verbose=1\n",
        ") , ModelCheckpoint('best_model.h5', monitor='val_loss', save_best_only=True, mode='min',verbose=1) ]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "18zVU25xGHLX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 595
        },
        "outputId": "7d6b34dc-6c3e-43fb-cb9f-3697591eed26"
      },
      "source": [
        "m.compile(optimizer=\"adam\",\n",
        "                loss=tf.losses.binary_crossentropy,\n",
        "                metrics=[\"accuracy\"])\n",
        "m.fit(x_train, y_train, validation_data=(x_val, y_val) ,callbacks = keras_callbacks ,epochs=20)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.5098 - accuracy: 0.7169 \n",
            "Epoch 00001: val_loss improved from inf to 0.61949, saving model to best_model.h5\n",
            "93/93 [==============================] - 1426s 15s/step - loss: 0.5098 - accuracy: 0.7169 - val_loss: 0.6195 - val_accuracy: 0.4607\n",
            "Epoch 2/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.2954 - accuracy: 0.8886 \n",
            "Epoch 00002: val_loss improved from 0.61949 to 0.45808, saving model to best_model.h5\n",
            "93/93 [==============================] - 1405s 15s/step - loss: 0.2954 - accuracy: 0.8886 - val_loss: 0.4581 - val_accuracy: 0.7579\n",
            "Epoch 3/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.1919 - accuracy: 0.9347 \n",
            "Epoch 00003: val_loss improved from 0.45808 to 0.18941, saving model to best_model.h5\n",
            "93/93 [==============================] - 1424s 15s/step - loss: 0.1919 - accuracy: 0.9347 - val_loss: 0.1894 - val_accuracy: 0.9292\n",
            "Epoch 4/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.1171 - accuracy: 0.9643 \n",
            "Epoch 00004: val_loss did not improve from 0.18941\n",
            "93/93 [==============================] - 1462s 16s/step - loss: 0.1171 - accuracy: 0.9643 - val_loss: 0.3775 - val_accuracy: 0.8679\n",
            "Epoch 5/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.0899 - accuracy: 0.9711 \n",
            "Epoch 00005: val_loss did not improve from 0.18941\n",
            "93/93 [==============================] - 1452s 16s/step - loss: 0.0899 - accuracy: 0.9711 - val_loss: 0.2794 - val_accuracy: 0.8962\n",
            "Epoch 6/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.0617 - accuracy: 0.9812 \n",
            "Epoch 00006: val_loss did not improve from 0.18941\n",
            "93/93 [==============================] - 1446s 16s/step - loss: 0.0617 - accuracy: 0.9812 - val_loss: 0.4354 - val_accuracy: 0.8255\n",
            "Epoch 7/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.0536 - accuracy: 0.9825 \n",
            "Epoch 00007: val_loss did not improve from 0.18941\n",
            "93/93 [==============================] - 1442s 16s/step - loss: 0.0536 - accuracy: 0.9825 - val_loss: 0.2679 - val_accuracy: 0.9119\n",
            "Epoch 8/20\n",
            "93/93 [==============================] - ETA: 0s - loss: 0.0512 - accuracy: 0.9785 \n",
            "Epoch 00008: val_loss did not improve from 0.18941\n",
            "93/93 [==============================] - 1446s 16s/step - loss: 0.0512 - accuracy: 0.9785 - val_loss: 0.7195 - val_accuracy: 0.8758\n",
            "Epoch 00008: early stopping\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f5552f6ab70>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AUgIn0h8Tp62",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "52a17bfd-e251-485a-ecc1-77ae8e29131b"
      },
      "source": [
        "test_loss, test_acc = m.evaluate(x_test, y_test)\n",
        "print(\"Loss: {0} - Acc: {1}\".format(test_loss, test_acc))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "20/20 [==============================] - 75s 4s/step - loss: 2.3038 - accuracy: 0.4969\n",
            "Loss: 2.3037517070770264 - Acc: 0.49685534834861755\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L5j-Ni-vGHIY",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 231
        },
        "outputId": "af26a021-a07e-4e94-e2c9-4dc4c7901bfc"
      },
      "source": [
        "epochs = 8\n",
        "epoch_range = range(1, epochs+1)\n",
        "plt.plot(epoch_range, history.history['accuracy'])\n",
        "plt.plot(epoch_range, history.history['val_accuracy'])\n",
        "plt.title('Model accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "# Plot training & validation loss values\n",
        "plt.plot(epoch_range, history.history['loss'])\n",
        "plt.plot(epoch_range, history.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-46-600cd8375097>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mepochs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m8\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mepoch_range\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mepoch_range\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhistory\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'accuracy'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mepoch_range\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhistory\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'val_accuracy'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtitle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Model accuracy'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'history' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uJX7y9wgAvi0"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qwRnumMdBTwW"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NFT3gbDdCE3u"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4OsMHUZxpbwr"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    }
  ]
}