Preparing_a_TensorFlow_model_for_a_mobile_device_with_TensorFlow_Lite.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Preparing a TensorFlow model for a mobile device with TensorFlow Lite.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "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/sergejhorvat/Tensorflow2.0_Udemy/blob/master/Preparing_a_TensorFlow_model_for_a_mobile_device_with_TensorFlow_Lite.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gNCP2RT3RNtz",
        "colab_type": "text"
      },
      "source": [
        "## Stage 1: Installing dependencies and setting up the environment"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OD7YjsLfdayw",
        "colab_type": "text"
      },
      "source": [
        "Using TF 1.13.1 becaouse in TF 2 there is not yet TF Lite library."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RtAydW1fQ2Kg",
        "colab_type": "code",
        "outputId": "7f2bb770-bdbc-446c-eaca-6782b66b8b05",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 698
        }
      },
      "source": [
        "!pip install tensorflow-gpu==1.13.1"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting tensorflow-gpu==1.13.1\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/7b/b1/0ad4ae02e17ddd62109cd54c291e311c4b5fd09b4d0678d3d6ce4159b0f0/tensorflow_gpu-1.13.1-cp36-cp36m-manylinux1_x86_64.whl (345.2MB)\n",
            "\u001b[K     |████████████████████████████████| 345.2MB 79kB/s \n",
            "\u001b[?25hRequirement already satisfied: gast>=0.2.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (0.3.2)\n",
            "Requirement already satisfied: keras-preprocessing>=1.0.5 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (1.1.0)\n",
            "Requirement already satisfied: grpcio>=1.8.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (1.15.0)\n",
            "Requirement already satisfied: wheel>=0.26 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (0.33.6)\n",
            "Requirement already satisfied: protobuf>=3.6.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (3.7.1)\n",
            "Requirement already satisfied: absl-py>=0.1.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (0.8.0)\n",
            "Requirement already satisfied: astor>=0.6.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (0.8.0)\n",
            "Requirement already satisfied: numpy>=1.13.3 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (1.16.5)\n",
            "Collecting tensorboard<1.14.0,>=1.13.0 (from tensorflow-gpu==1.13.1)\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/0f/39/bdd75b08a6fba41f098b6cb091b9e8c7a80e1b4d679a581a0ccd17b10373/tensorboard-1.13.1-py3-none-any.whl (3.2MB)\n",
            "\u001b[K     |████████████████████████████████| 3.2MB 42.6MB/s \n",
            "\u001b[?25hRequirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (1.1.0)\n",
            "Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (1.12.0)\n",
            "Collecting tensorflow-estimator<1.14.0rc0,>=1.13.0 (from tensorflow-gpu==1.13.1)\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/bb/48/13f49fc3fa0fdf916aa1419013bb8f2ad09674c275b4046d5ee669a46873/tensorflow_estimator-1.13.0-py2.py3-none-any.whl (367kB)\n",
            "\u001b[K     |████████████████████████████████| 368kB 46.3MB/s \n",
            "\u001b[?25hRequirement already satisfied: keras-applications>=1.0.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.13.1) (1.0.8)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from protobuf>=3.6.1->tensorflow-gpu==1.13.1) (41.2.0)\n",
            "Requirement already satisfied: werkzeug>=0.11.15 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.14.0,>=1.13.0->tensorflow-gpu==1.13.1) (0.16.0)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.14.0,>=1.13.0->tensorflow-gpu==1.13.1) (3.1.1)\n",
            "Collecting mock>=2.0.0 (from tensorflow-estimator<1.14.0rc0,>=1.13.0->tensorflow-gpu==1.13.1)\n",
            "  Downloading https://files.pythonhosted.org/packages/05/d2/f94e68be6b17f46d2c353564da56e6fb89ef09faeeff3313a046cb810ca9/mock-3.0.5-py2.py3-none-any.whl\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.6/dist-packages (from keras-applications>=1.0.6->tensorflow-gpu==1.13.1) (2.8.0)\n",
            "\u001b[31mERROR: tensorflow 1.14.0 has requirement tensorboard<1.15.0,>=1.14.0, but you'll have tensorboard 1.13.1 which is incompatible.\u001b[0m\n",
            "\u001b[31mERROR: tensorflow 1.14.0 has requirement tensorflow-estimator<1.15.0rc0,>=1.14.0rc0, but you'll have tensorflow-estimator 1.13.0 which is incompatible.\u001b[0m\n",
            "Installing collected packages: tensorboard, mock, tensorflow-estimator, tensorflow-gpu\n",
            "  Found existing installation: tensorboard 1.14.0\n",
            "    Uninstalling tensorboard-1.14.0:\n",
            "      Successfully uninstalled tensorboard-1.14.0\n",
            "  Found existing installation: tensorflow-estimator 1.14.0\n",
            "    Uninstalling tensorflow-estimator-1.14.0:\n",
            "      Successfully uninstalled tensorflow-estimator-1.14.0\n",
            "Successfully installed mock-3.0.5 tensorboard-1.13.1 tensorflow-estimator-1.13.0 tensorflow-gpu-1.13.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i7guNG2NSYm9",
        "colab_type": "text"
      },
      "source": [
        "## Stage 2: Importing project dependencies"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p3kH8SQSSA6c",
        "colab_type": "code",
        "outputId": "ebc1eea8-0604-4a71-d8f2-2e4f25f63c6e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "\n",
        "from tensorflow.keras.datasets import fashion_mnist\n",
        "\n",
        "tf.__version__"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'1.13.1'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qUemNsucSpTc",
        "colab_type": "text"
      },
      "source": [
        "## Stage 3: Dataset preprocessing"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NTxQSWirSs2l",
        "colab_type": "text"
      },
      "source": [
        "### Loading the FashionMNIST dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3s83WwJ2SnDr",
        "colab_type": "code",
        "outputId": "9d52bddc-dce5-4d5c-b475-9a2fdcbc24d1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 184
        }
      },
      "source": [
        "(X_train, y_train), (X_test, y_test) = fashion_mnist.load_data()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz\n",
            "32768/29515 [=================================] - 0s 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz\n",
            "26427392/26421880 [==============================] - 0s 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz\n",
            "8192/5148 [===============================================] - 0s 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz\n",
            "4423680/4422102 [==============================] - 0s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4OvP_vFiSwmc",
        "colab_type": "text"
      },
      "source": [
        "### Image normalization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rODvuie6Sz06",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X_train = X_train / 255.\n",
        "X_test = X_test / 255."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hky7XzH3S1sQ",
        "colab_type": "text"
      },
      "source": [
        "### Reshaping of the dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aBxZsTPBS4XJ",
        "colab_type": "code",
        "outputId": "48a5615f-c56d-4b7c-83a1-e48527e6493e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "X_train.shape"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(60000, 28, 28)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5rPdViS9T1Xb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# -1 takes all images in array\n",
        "X_train = X_train.reshape(-1, 28*28)\n",
        "X_test = X_test.reshape(-1, 28*28)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fbmf82l2UH3a",
        "colab_type": "code",
        "outputId": "807a2549-cca7-4899-df34-c651bcaaff5b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "X_train.shape"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(60000, 784)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "o7g7SeZgUWCo",
        "colab_type": "text"
      },
      "source": [
        "## Stage 4: Building a model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HfqQPaYVUY2S",
        "colab_type": "text"
      },
      "source": [
        "### Defining the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "db7lvvFXUKe6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = tf.keras.models.Sequential()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RrwtIwtCUa0g",
        "colab_type": "text"
      },
      "source": [
        "### Building the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pnSHfe5XUcp1",
        "colab_type": "code",
        "outputId": "5328763e-d9dc-4448-9cd5-c39f10559614",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 147
        }
      },
      "source": [
        "model.add(tf.keras.layers.Dense(units=128, activation='relu', input_shape=(784,)))\n",
        "model.add(tf.keras.layers.Dropout(rate=0.2))\n",
        "model.add(tf.keras.layers.Dense(units=10, activation='softmax'))"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Colocations handled automatically by placer.\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/layers/core.py:143: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ed_YGh-5UdL3",
        "colab_type": "text"
      },
      "source": [
        "### Compiling the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JZVOTYQXUgC5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZnV-v8ooWXPw",
        "colab_type": "text"
      },
      "source": [
        "### Training the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oZnrU1MHVB4b",
        "colab_type": "code",
        "outputId": "3818d887-9720-4bfc-9429-96c1c8fdec91",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 239
        }
      },
      "source": [
        "model.fit(X_train, y_train, epochs=5)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/5\n",
            "60000/60000 [==============================] - 8s 136us/sample - loss: 0.5318 - sparse_categorical_accuracy: 0.8140\n",
            "Epoch 2/5\n",
            "60000/60000 [==============================] - 7s 115us/sample - loss: 0.3997 - sparse_categorical_accuracy: 0.8555\n",
            "Epoch 3/5\n",
            "60000/60000 [==============================] - 7s 114us/sample - loss: 0.3692 - sparse_categorical_accuracy: 0.8656\n",
            "Epoch 4/5\n",
            "60000/60000 [==============================] - 7s 116us/sample - loss: 0.3457 - sparse_categorical_accuracy: 0.8730\n",
            "Epoch 5/5\n",
            "60000/60000 [==============================] - 7s 116us/sample - loss: 0.3271 - sparse_categorical_accuracy: 0.8795\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7fba8a2d4518>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P2GalF1WWZDk",
        "colab_type": "text"
      },
      "source": [
        "### Evaluating the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zUO2saLWWa3q",
        "colab_type": "code",
        "outputId": "f110b4d1-db41-43ea-8c64-f8335857acaf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 74
        }
      },
      "source": [
        "test_loss, test_accuracy = model.evaluate(X_test, y_test)\n",
        "print(\"Test accuracy: {}\".format(test_accuracy))"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "10000/10000 [==============================] - 1s 64us/sample - loss: 0.3622 - sparse_categorical_accuracy: 0.8703\n",
            "Test accuracy: 0.8702999949455261\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AV_Vnn1IXbUN",
        "colab_type": "text"
      },
      "source": [
        "## Stage 5: Converting the model into TensorFlow Lite version"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bw6WoYDeXhA4",
        "colab_type": "text"
      },
      "source": [
        "### Saving the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Lu3wsEWAXMFC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_name = 'fashion_mobile_model.h5'\n",
        "tf.keras.models.save_model(model, model_name)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Bvz3KYQWYKe2",
        "colab_type": "text"
      },
      "source": [
        "### Creating the TFLite Converter"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7mgW1OGcYp5b",
        "colab_type": "code",
        "outputId": "0b3bfada-2009-4747-800c-bc468b12a731",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 184
        }
      },
      "source": [
        "converter = tf.lite.TFLiteConverter.from_keras_model_file(model_name)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/lite/python/lite.py:591: convert_variables_to_constants (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.compat.v1.graph_util.convert_variables_to_constants\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/graph_util_impl.py:245: extract_sub_graph (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.compat.v1.graph_util.extract_sub_graph\n",
            "INFO:tensorflow:Froze 4 variables.\n",
            "INFO:tensorflow:Converted 4 variables to const ops.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R2kkEO_bZbcI",
        "colab_type": "text"
      },
      "source": [
        "### Converting the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-TpOJV5mZTRC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "tflite_model = converter.convert()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dXHT978hZzX3",
        "colab_type": "text"
      },
      "source": [
        "### Saving the TFLite version of the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9jqZ2zpmZnez",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "with open(\"tf_model.tflite\", \"wb\") as f:\n",
        "  f.write(tflite_model)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7tMRuHHXuvMp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}