ResNet_Transfer_Learning_TensorFlow.ipynb

resnet_transfer_learning_tensorflow.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ResNet_Transfer_Learning_TensorFlow.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyPkF5cbObNA2gzRa9KkWXrw",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/mrgrhn/b2eece6b41d9f21657a97c8367d52dd0/resnet_transfer_learning_tensorflow.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KXQu67zVoZAw"
      },
      "source": [
        "import tensorflow as tf\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "from tensorflow.keras import datasets, layers, models, losses, Model"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fg4d0i01ofqp",
        "outputId": "eb3adca9-b977-4593-a5df-b2fb8e9d3de1"
      },
      "source": [
        "(x_train, y_train), (x_test, y_test)=tf.keras.datasets.mnist.load_data()\r\n",
        "x_train = tf.pad(x_train, [[0, 0], [2,2], [2,2]])/255\r\n",
        "x_test = tf.pad(x_test, [[0, 0], [2,2], [2,2]])/255\r\n",
        "x_train = tf.expand_dims(x_train, axis=3, name=None)\r\n",
        "x_test = tf.expand_dims(x_test, axis=3, name=None)\r\n",
        "x_train = tf.repeat(x_train, 3, axis=3)\r\n",
        "x_test = tf.repeat(x_test, 3, axis=3)\r\n",
        "x_val = x_train[-2000:,:,:]\r\n",
        "y_val = y_train[-2000:]\r\n",
        "x_train = x_train[:-2000,:,:]\r\n",
        "y_train = y_train[:-2000]"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d2mRBkTBoig8",
        "outputId": "7aeee00a-bdff-41db-8473-6e04850605b5"
      },
      "source": [
        "base_model = tf.keras.applications.ResNet152(weights = 'imagenet', include_top = False, input_shape = (32,32,3))\r\n",
        "for layer in base_model.layers:\r\n",
        "  layer.trainable = False"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet152_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "234700800/234698864 [==============================] - 4s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rWSL-tt4qBRa"
      },
      "source": [
        "x = layers.Flatten()(base_model.output)\r\n",
        "x = layers.Dense(1000, activation='relu')(x)\r\n",
        "predictions = layers.Dense(10, activation = 'softmax')(x)"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s1sYr8cIt0nY"
      },
      "source": [
        "head_model = Model(inputs = base_model.input, outputs = predictions)\r\n",
        "head_model.compile(optimizer='adam', loss=losses.sparse_categorical_crossentropy, metrics=['accuracy'])"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BKH1F-put3oU",
        "outputId": "18bbef30-4a2c-4bde-9807-5afe16eb62fe"
      },
      "source": [
        "history = head_model.fit(x_train, y_train, batch_size=64, epochs=40, validation_data=(x_val, y_val))"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/40\n",
            "907/907 [==============================] - 44s 33ms/step - loss: 1.2104 - accuracy: 0.5902 - val_loss: 0.4982 - val_accuracy: 0.8450\n",
            "Epoch 2/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.6148 - accuracy: 0.8014 - val_loss: 0.3869 - val_accuracy: 0.8955\n",
            "Epoch 3/40\n",
            "907/907 [==============================] - 29s 32ms/step - loss: 0.5172 - accuracy: 0.8325 - val_loss: 0.3259 - val_accuracy: 0.9065\n",
            "Epoch 4/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.4702 - accuracy: 0.8459 - val_loss: 0.3525 - val_accuracy: 0.8925\n",
            "Epoch 5/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.4441 - accuracy: 0.8518 - val_loss: 0.3125 - val_accuracy: 0.9125\n",
            "Epoch 6/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.4011 - accuracy: 0.8679 - val_loss: 0.2554 - val_accuracy: 0.9310\n",
            "Epoch 7/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.3739 - accuracy: 0.8790 - val_loss: 0.3459 - val_accuracy: 0.8920\n",
            "Epoch 8/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.3746 - accuracy: 0.8763 - val_loss: 0.2262 - val_accuracy: 0.9430\n",
            "Epoch 9/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.3535 - accuracy: 0.8820 - val_loss: 0.2385 - val_accuracy: 0.9315\n",
            "Epoch 10/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.3262 - accuracy: 0.8922 - val_loss: 0.2961 - val_accuracy: 0.9130\n",
            "Epoch 11/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.3333 - accuracy: 0.8897 - val_loss: 0.2265 - val_accuracy: 0.9350\n",
            "Epoch 12/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.3168 - accuracy: 0.8948 - val_loss: 0.2031 - val_accuracy: 0.9485\n",
            "Epoch 13/40\n",
            "907/907 [==============================] - 29s 32ms/step - loss: 0.3160 - accuracy: 0.8949 - val_loss: 0.2349 - val_accuracy: 0.9335\n",
            "Epoch 14/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.3033 - accuracy: 0.8997 - val_loss: 0.2310 - val_accuracy: 0.9265\n",
            "Epoch 15/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.3025 - accuracy: 0.9005 - val_loss: 0.2259 - val_accuracy: 0.9355\n",
            "Epoch 16/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.2906 - accuracy: 0.9054 - val_loss: 0.2388 - val_accuracy: 0.9240\n",
            "Epoch 17/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.2891 - accuracy: 0.9050 - val_loss: 0.2096 - val_accuracy: 0.9380\n",
            "Epoch 18/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2774 - accuracy: 0.9094 - val_loss: 0.1940 - val_accuracy: 0.9445\n",
            "Epoch 19/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2626 - accuracy: 0.9137 - val_loss: 0.1886 - val_accuracy: 0.9420\n",
            "Epoch 20/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.2654 - accuracy: 0.9114 - val_loss: 0.1828 - val_accuracy: 0.9435\n",
            "Epoch 21/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2580 - accuracy: 0.9158 - val_loss: 0.1988 - val_accuracy: 0.9400\n",
            "Epoch 22/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2605 - accuracy: 0.9154 - val_loss: 0.1901 - val_accuracy: 0.9465\n",
            "Epoch 23/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.2594 - accuracy: 0.9134 - val_loss: 0.1846 - val_accuracy: 0.9465\n",
            "Epoch 24/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2466 - accuracy: 0.9173 - val_loss: 0.1702 - val_accuracy: 0.9515\n",
            "Epoch 25/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2460 - accuracy: 0.9192 - val_loss: 0.2128 - val_accuracy: 0.9375\n",
            "Epoch 26/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2480 - accuracy: 0.9187 - val_loss: 0.1885 - val_accuracy: 0.9470\n",
            "Epoch 27/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2399 - accuracy: 0.9210 - val_loss: 0.1786 - val_accuracy: 0.9445\n",
            "Epoch 28/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2433 - accuracy: 0.9202 - val_loss: 0.1664 - val_accuracy: 0.9555\n",
            "Epoch 29/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2382 - accuracy: 0.9223 - val_loss: 0.1720 - val_accuracy: 0.9460\n",
            "Epoch 30/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.2312 - accuracy: 0.9234 - val_loss: 0.1706 - val_accuracy: 0.9505\n",
            "Epoch 31/40\n",
            "907/907 [==============================] - 29s 31ms/step - loss: 0.2340 - accuracy: 0.9236 - val_loss: 0.1706 - val_accuracy: 0.9525\n",
            "Epoch 32/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2327 - accuracy: 0.9245 - val_loss: 0.1649 - val_accuracy: 0.9485\n",
            "Epoch 33/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2275 - accuracy: 0.9262 - val_loss: 0.1599 - val_accuracy: 0.9505\n",
            "Epoch 34/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2274 - accuracy: 0.9259 - val_loss: 0.1555 - val_accuracy: 0.9550\n",
            "Epoch 35/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2217 - accuracy: 0.9273 - val_loss: 0.1604 - val_accuracy: 0.9530\n",
            "Epoch 36/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2197 - accuracy: 0.9260 - val_loss: 0.1702 - val_accuracy: 0.9520\n",
            "Epoch 37/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2228 - accuracy: 0.9268 - val_loss: 0.1843 - val_accuracy: 0.9450\n",
            "Epoch 38/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2173 - accuracy: 0.9283 - val_loss: 0.1729 - val_accuracy: 0.9495\n",
            "Epoch 39/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2081 - accuracy: 0.9313 - val_loss: 0.1600 - val_accuracy: 0.9535\n",
            "Epoch 40/40\n",
            "907/907 [==============================] - 28s 31ms/step - loss: 0.2082 - accuracy: 0.9326 - val_loss: 0.1713 - val_accuracy: 0.9495\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 910
        },
        "id": "yE3OUgo34v_C",
        "outputId": "fd989d09-d97f-478e-e768-0e2e739a17aa"
      },
      "source": [
        "fig, axs = plt.subplots(2, 1, figsize=(15,15))\r\n",
        "\r\n",
        "axs[0].plot(history.history['loss'])\r\n",
        "axs[0].plot(history.history['val_loss'])\r\n",
        "axs[0].title.set_text('Training Loss vs Validation Loss')\r\n",
        "axs[0].set_xlabel('Epochs')\r\n",
        "axs[0].set_ylabel('Loss')\r\n",
        "axs[0].legend(['Train','Val'])\r\n",
        "\r\n",
        "axs[1].plot(history.history['accuracy'])\r\n",
        "axs[1].plot(history.history['val_accuracy'])\r\n",
        "axs[1].title.set_text('Training Accuracy vs Validation Accuracy')\r\n",
        "axs[1].set_xlabel('Epochs')\r\n",
        "axs[1].set_ylabel('Accuracy')\r\n",
        "axs[1].legend(['Train', 'Val'])"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f54150d2be0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x1080 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3b5bRktf4ydg",
        "outputId": "37f5eb06-7caa-4c5f-b6fd-9b5758a1f057"
      },
      "source": [
        "head_model.evaluate(x_test, y_test)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 7s 20ms/step - loss: 0.2287 - accuracy: 0.9275\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.22868101298809052, 0.9275000095367432]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    }
  ]
}