TF_Hub_for_TF2_Image_Module_Retraining_(preview).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "TF Hub for TF2: Image Module Retraining (preview)",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [
        "ScitaPqhKtuW"
      ],
      "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/github/ActionPace/DeepLearningTests/blob/master/TF_Hub_for_TF2_Image_Module_Retraining_(preview).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "ScitaPqhKtuW"
      },
      "cell_type": "markdown",
      "source": [
        "##### Copyright 2019 The TensorFlow Hub Authors.\n",
        "\n",
        "Licensed under the Apache License, Version 2.0 (the \"License\");"
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "jvztxQ6VsK2k",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Copyright 2019 The TensorFlow Hub Authors. All Rights Reserved.\n",
        "#\n",
        "# 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",
        "#     http://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.\n",
        "# =============================================================================="
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "oYM61xrTsP5d"
      },
      "cell_type": "markdown",
      "source": [
        "# TF Hub for TF2: Image Module Retraining (preview)\n",
        "\n",
        "<table align=\"left\">\n",
        "<td align=\"center\">\n",
        "  <a target=\"_blank\"  href=\"https://colab.research.google.com/github/tensorflow/hub/blob/master/examples/colab/tf2_image_retraining.ipynb\">\n",
        "    <img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" /><br>Run in Google Colab\n",
        "  </a>\n",
        "</td>\n",
        "<td align=\"center\">\n",
        "  <a target=\"_blank\"  href=\"https://github.com/tensorflow/hub/blob/master/examples/colab/tf2_image_retraining.ipynb\">\n",
        "    <img width=32px src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" /><br>View source on GitHub</a>\n",
        "</td>\n",
        "</table>"
      ]
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "L1otmJgmbahf"
      },
      "cell_type": "markdown",
      "source": [
        "This Colab offers a preview of using TensorFlow Hub modules in the native TF2 format with Keras. It uses a pre-trained image feature vector module for classifying five species of flowers, including fine-tuning of the module.\n",
        "\n",
        "**NOTE:** No stable versions of TF2 and TF Hub for TF2 have been released at this point. This colab needs PIP package `tensorflow-hub>=0.3.0` to run."
      ]
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "bL54LWCHt5q5"
      },
      "cell_type": "markdown",
      "source": [
        "## Set up library versions for TF2"
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "110fGB18UNJn",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 69
        },
        "outputId": "408b33ed-382e-4388-fe54-feedfbebb016"
      },
      "cell_type": "code",
      "source": [
        "!pip install -U --pre tensorflow-gpu --quiet"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\u001b[K    100% |████████████████████████████████| 332.1MB 50kB/s \n",
            "\u001b[K    100% |████████████████████████████████| 419kB 13.0MB/s \n",
            "\u001b[K    100% |████████████████████████████████| 3.0MB 1.1MB/s \n",
            "\u001b[?25h"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "FewOCz59V-bd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 104
        },
        "outputId": "12347685-3c03-42b6-83ff-c1101f4c3b25"
      },
      "cell_type": "code",
      "source": [
        "!pip install 'tensorflow-hub>=0.3'"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: tensorflow-hub>=0.3 in /usr/local/lib/python3.6/dist-packages (0.4.0)\n",
            "Requirement already satisfied: protobuf>=3.4.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-hub>=0.3) (3.7.1)\n",
            "Requirement already satisfied: numpy>=1.12.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-hub>=0.3) (1.16.2)\n",
            "Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-hub>=0.3) (1.11.0)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from protobuf>=3.4.0->tensorflow-hub>=0.3) (40.9.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "dlauq-4FWGZM",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        },
        "outputId": "a3c8da22-9380-4184-c1b1-081c5b457dd0"
      },
      "cell_type": "code",
      "source": [
        "from __future__ import absolute_import, division, print_function\n",
        "\n",
        "import os\n",
        "\n",
        "import matplotlib.pylab as plt\n",
        "import numpy as np\n",
        "\n",
        "import tensorflow as tf\n",
        "import tensorflow_hub as hub\n",
        "\n",
        "print(\"Version: \", tf.__version__)\n",
        "print(\"Eager mode: \", tf.executing_eagerly())\n",
        "print(\"Hub version: \", hub.__version__)\n",
        "print(\"GPU is\", \"available\" if tf.test.is_gpu_available() else \"NOT AVAILABLE\")"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Version:  2.0.0-alpha0\n",
            "Eager mode:  True\n",
            "Hub version:  0.4.0\n",
            "GPU is available\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "mmaHHH7Pvmth"
      },
      "cell_type": "markdown",
      "source": [
        "## Select the Hub/TF2 module to use\n",
        "\n",
        "Hub modules for TF 1.x won't work here, please use one of the selections provided."
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "FlsEcKVeuCnf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "6da25ac5-d914-4765-c86f-3c11c2632b7e"
      },
      "cell_type": "code",
      "source": [
        "module_selection = (\"mobilenet_v2\", 224, 1280) #@param [\"(\\\"mobilenet_v2\\\", 224, 1280)\", \"(\\\"inception_v3\\\", 299, 2048)\"] {type:\"raw\", allow-input: true}\n",
        "handle_base, pixels, FV_SIZE = module_selection\n",
        "MODULE_HANDLE =\"https://tfhub.dev/google/tf2-preview/{}/feature_vector/2\".format(handle_base)\n",
        "IMAGE_SIZE = (pixels, pixels)\n",
        "print(\"Using {} with input size {} and output dimension {}\".format(\n",
        "  MODULE_HANDLE, IMAGE_SIZE, FV_SIZE))\n",
        "\n",
        "BATCH_SIZE = 32 #@param {type:\"integer\"}"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using https://tfhub.dev/google/tf2-preview/mobilenet_v2/feature_vector/2 with input size (224, 224) and output dimension 1280\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "yTY8qzyYv3vl"
      },
      "cell_type": "markdown",
      "source": [
        "## Set up the Flowers dataset\n",
        "\n",
        "Inputs are suitably resized for the selected module. Dataset augmentation (i.e., random distortions of an image each time it is read) improves training, esp. when fine-tuning."
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "WBtFK1hO8KsO",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "data_dir = tf.keras.utils.get_file(\n",
        "    'flower_photos',\n",
        "    'https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz',\n",
        "    untar=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "BzysZO58YA8H",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "495b947f-112d-46e3-e471-aa8d8592086c"
      },
      "cell_type": "code",
      "source": [
        "datagen_kwargs = dict(rescale=1./255, validation_split=.20)\n",
        "valid_datagen = tf.keras.preprocessing.image.ImageDataGenerator(\n",
        "    **datagen_kwargs)\n",
        "valid_generator = valid_datagen.flow_from_directory(\n",
        "    data_dir, subset=\"validation\", shuffle=False,\n",
        "    target_size=IMAGE_SIZE, batch_size=BATCH_SIZE)\n",
        "\n",
        "do_data_augmentation = False #@param {type:\"boolean\"}\n",
        "if do_data_augmentation:\n",
        "  train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(\n",
        "      rotation_range=40,\n",
        "      horizontal_flip=True,\n",
        "      width_shift_range=0.2, height_shift_range=0.2,\n",
        "      shear_range=0.2, zoom_range=0.2,\n",
        "      **datagen_kwargs)\n",
        "else:\n",
        "  train_datagen = valid_datagen\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "    data_dir, subset=\"training\", shuffle=True,\n",
        "    target_size=IMAGE_SIZE, batch_size=BATCH_SIZE)\n"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 731 images belonging to 5 classes.\n",
            "Found 2939 images belonging to 5 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "FS_gVStowW3G"
      },
      "cell_type": "markdown",
      "source": [
        "\n",
        "## Defining the model\n",
        "\n",
        "All it takes is to put a linear classifier on top of the `feature_extractor_layer` with the Hub module.\n",
        "\n",
        "For speed, we start out with a non-trainable `feature_extractor_layer`, but you can also enable fine-tuning for greater accuracy."
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "RaJW3XrPyFiF",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "do_fine_tuning = False #@param {type:\"boolean\"}"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "50FYNIb1dmJH",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 277
        },
        "outputId": "86d45671-73df-4a12-997a-db10d8155c61"
      },
      "cell_type": "code",
      "source": [
        "print(\"Building model with\", MODULE_HANDLE)\n",
        "model = tf.keras.Sequential([\n",
        "    hub.KerasLayer(MODULE_HANDLE, output_shape=[FV_SIZE],\n",
        "                   trainable=do_fine_tuning),\n",
        "    tf.keras.layers.Dropout(rate=0.2),\n",
        "    tf.keras.layers.Dense(train_generator.num_classes, activation='softmax',\n",
        "                          kernel_regularizer=tf.keras.regularizers.l2(0.0001))\n",
        "])\n",
        "model.build((None,)+IMAGE_SIZE+(3,))\n",
        "model.summary()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Building model with https://tfhub.dev/google/tf2-preview/mobilenet_v2/feature_vector/2\n",
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "keras_layer (KerasLayer)     multiple                  2257984   \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            multiple                  0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                multiple                  6405      \n",
            "=================================================================\n",
            "Total params: 2,264,389\n",
            "Trainable params: 6,405\n",
            "Non-trainable params: 2,257,984\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "colab_type": "text",
        "id": "u2e5WupIw2N2"
      },
      "cell_type": "markdown",
      "source": [
        "## Training the model"
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "9f3yBUvkd_VJ",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model.compile(\n",
        "  optimizer=tf.keras.optimizers.SGD(lr=0.005, momentum=0.9), \n",
        "  loss=tf.keras.losses.CategoricalCrossentropy(label_smoothing=0.1),\n",
        "  metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "w_YKX2Qnfg6x",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 191
        },
        "outputId": "4ad0ccd7-ac38-4042-9659-447426ec64ea"
      },
      "cell_type": "code",
      "source": [
        "steps_per_epoch = train_generator.samples // train_generator.batch_size\n",
        "validation_steps = valid_generator.samples // valid_generator.batch_size\n",
        "hist = model.fit_generator(\n",
        "    train_generator,\n",
        "    epochs=5, steps_per_epoch=steps_per_epoch,\n",
        "    validation_data=valid_generator,\n",
        "    validation_steps=validation_steps).history"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/5\n",
            "91/91 [==============================] - 22s 245ms/step - loss: 1.0975 - accuracy: 0.5446 - val_loss: 0.8640 - val_accuracy: 0.8295\n",
            "Epoch 2/5\n",
            "91/91 [==============================] - 13s 147ms/step - loss: 0.8407 - accuracy: 0.8280 - val_loss: 0.8257 - val_accuracy: 0.8523\n",
            "Epoch 3/5\n",
            "91/91 [==============================] - 14s 153ms/step - loss: 0.8257 - accuracy: 0.8427 - val_loss: 0.8165 - val_accuracy: 0.8636\n",
            "Epoch 4/5\n",
            "91/91 [==============================] - 14s 149ms/step - loss: 0.7804 - accuracy: 0.8630 - val_loss: 0.8026 - val_accuracy: 0.8608\n",
            "Epoch 5/5\n",
            "91/91 [==============================] - 14s 158ms/step - loss: 0.7628 - accuracy: 0.8625 - val_loss: 0.8124 - val_accuracy: 0.8565\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "colab_type": "code",
        "id": "CYOw0fTO1W4x",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 567
        },
        "outputId": "bb44d240-2d48-4151-b9fe-4cb766e6aaf2"
      },
      "cell_type": "code",
      "source": [
        "plt.figure()\n",
        "plt.ylabel(\"Loss (training and validation)\")\n",
        "plt.xlabel(\"Training Steps\")\n",
        "plt.ylim([0,2])\n",
        "plt.plot(hist[\"loss\"])\n",
        "plt.plot(hist[\"val_loss\"])\n",
        "\n",
        "plt.figure()\n",
        "plt.ylabel(\"Accuracy (training and validation)\")\n",
        "plt.xlabel(\"Training Steps\")\n",
        "plt.ylim([0,1])\n",
        "plt.plot(hist[\"accuracy\"])\n",
        "plt.plot(hist[\"val_accuracy\"])"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f1aa65d4080>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEKCAYAAAAB0GKPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEKCAYAAAD9xUlFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQAAE6OuNvAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "TXQQFLLoEUA3",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}