cifar10_training.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jndmk1TVPUyp"
   },
   "source": [
    "# CIFAR10 Training and quantization\n",
    "\n",
    "Project inspired by this [kaggle competition](https://www.kaggle.com/ektasharma/simple-cifar10-cnn-keras-code-with-88-accuracy#A-Simple-Keras-CNN-trained-on-CIFAR-10-dataset-with-over-88%-accuracy-(Without-Data-Augmentation). In this colab you will be driven through the training of a CNN for CIFAR10 classification task. The model is then exported for inference with tflite and quantized, prepared for the *GAPflow*."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Cj6DcUxZAOse"
   },
   "source": [
    "### Used python modules"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "KNb0mejGAHCV"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-11-13 16:50:12.857774: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2022-11-13 16:50:12.968761: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory\n",
      "2022-11-13 16:50:12.968780: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.\n",
      "2022-11-13 16:50:12.993657: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
      "2022-11-13 16:50:13.465280: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory\n",
      "2022-11-13 16:50:13.465323: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory\n",
      "2022-11-13 16:50:13.465329: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.\n"
     ]
    }
   ],
   "source": [
    "# from nntool.api import NNGraph\n",
    "from model_v1 import *\n",
    "from tqdm import tqdm\n",
    "import tensorflow.keras as keras\n",
    "import tensorflow_model_optimization as tfmot\n",
    "from tensorflow.keras.datasets import cifar10\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras import datasets, layers, models\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from tensorflow.keras import regularizers\n",
    "from tensorflow.keras.layers import Dense, Dropout, BatchNormalization\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pathlib\n",
    "import os\n",
    "import tempfile\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ynH1CrIeC7l9",
    "outputId": "27185c8d-5330-4655-8353-8ab49cc6a76d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /tmp/ipykernel_18734/926465965.py:2: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use `tf.config.list_physical_devices('GPU')` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-11-13 16:50:16.960419: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2022-11-13 16:50:16.961587: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory\n",
      "2022-11-13 16:50:16.961611: W tensorflow/stream_executor/cuda/cuda_driver.cc:263] failed call to cuInit: UNKNOWN ERROR (303)\n",
      "2022-11-13 16:50:16.961633: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (cioflanc-HP-ENVY-x360-Convertible-13-ay0xxx): /proc/driver/nvidia/version does not exist\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "('2.10.0', False)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "tf.__version__, tf.test.is_gpu_available()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DTon9QahBjz3"
   },
   "source": [
    "# Reading the CIFAR-10 dataset from Keras datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Y73NtaTwAJGn",
    "outputId": "26ce625b-21aa-4153-ebf2-1103585b6fe4"
   },
   "outputs": [],
   "source": [
    "(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()\n",
    "train_images, train_labels = train_images[:10000], train_labels[:10000]\n",
    "test_images, test_labels = test_images[:1000], test_labels[:1000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "qnpbZeX8Ab08",
    "outputId": "afca2f22-80e7-4153-ca25-7b8bef0ac7e9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10000, 32, 32, 3)\n",
      "(10000, 1)\n",
      "(1000, 32, 32, 3)\n",
      "(1000, 1)\n"
     ]
    }
   ],
   "source": [
    "# Checking the number of rows (records) and columns (features)\n",
    "print(train_images.shape)\n",
    "print(train_labels.shape)\n",
    "print(test_images.shape)\n",
    "print(test_labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "uuLdAMSkBsQL",
    "outputId": "26fcf4ee-1c97-46b2-8663-f9b32151da1c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1 2 3 4 5 6 7 8 9]\n",
      "[0 1 2 3 4 5 6 7 8 9]\n"
     ]
    }
   ],
   "source": [
    "# Checking the number of unique classes \n",
    "print(np.unique(train_labels))\n",
    "print(np.unique(test_labels))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "0YoOBr90BtWH"
   },
   "outputs": [],
   "source": [
    "# Creating a list of all the class labels\n",
    "class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n",
    "               'dog', 'frog', 'horse', 'ship', 'truck']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 589
    },
    "id": "JwRU0I-sBuh3",
    "outputId": "45c3aa84-2f34-4c06-ef18-b9e3c6ab4c0f"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 25 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizing some of the images from the training dataset\n",
    "plt.figure(figsize=[10,10])\n",
    "for i in range (25):    # for first 25 images\n",
    "    plt.subplot(5, 5, i+1)\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "    plt.grid(False)\n",
    "    plt.imshow(train_images[i], cmap=plt.cm.binary)\n",
    "    plt.xlabel(class_names[train_labels[i][0]])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mA408CROB8m3"
   },
   "source": [
    "# Data Preprocessing\n",
    "\n",
    "Convert input [0:255] to [-1:1] float\n",
    "\n",
    "Convert labels to one hot encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "2HS36TpJBvrK"
   },
   "outputs": [],
   "source": [
    "# Converting the pixels data to float type\n",
    "train_images = train_images.astype('float32')\n",
    "test_images = test_images.astype('float32')\n",
    "\n",
    "# Standardizing (255 is the total number of pixels an image can have)\n",
    "train_images = (train_images / 128) - 1.0\n",
    "test_images = (test_images / 128) - 1.0\n",
    "\n",
    "# One hot encoding the target class (labels)\n",
    "num_classes = 10\n",
    "train_labels = to_categorical(train_labels, num_classes)\n",
    "test_labels = to_categorical(test_labels, num_classes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Design"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 32, 32, 16)        1216      \n",
      "                                                                 \n",
      " batch_normalization (BatchN  (None, 32, 32, 16)       64        \n",
      " ormalization)                                                   \n",
      "                                                                 \n",
      " re_lu (ReLU)                (None, 32, 32, 16)        0         \n",
      "                                                                 \n",
      " depthwise_conv2d (Depthwise  (None, 16, 16, 16)       160       \n",
      " Conv2D)                                                         \n",
      "                                                                 \n",
      " batch_normalization_1 (Batc  (None, 16, 16, 16)       64        \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_1 (ReLU)              (None, 16, 16, 16)        0         \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 16, 16, 32)        544       \n",
      "                                                                 \n",
      " batch_normalization_2 (Batc  (None, 16, 16, 32)       128       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_2 (ReLU)              (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " depthwise_conv2d_1 (Depthwi  (None, 16, 16, 32)       320       \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_3 (Batc  (None, 16, 16, 32)       128       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_3 (ReLU)              (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 16, 16, 48)        1584      \n",
      "                                                                 \n",
      " batch_normalization_4 (Batc  (None, 16, 16, 48)       192       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_4 (ReLU)              (None, 16, 16, 48)        0         \n",
      "                                                                 \n",
      " depthwise_conv2d_2 (Depthwi  (None, 8, 8, 48)         480       \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_5 (Batc  (None, 8, 8, 48)         192       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_5 (ReLU)              (None, 8, 8, 48)          0         \n",
      "                                                                 \n",
      " conv2d_3 (Conv2D)           (None, 8, 8, 64)          3136      \n",
      "                                                                 \n",
      " batch_normalization_6 (Batc  (None, 8, 8, 64)         256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_6 (ReLU)              (None, 8, 8, 64)          0         \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 8, 8, 64)          0         \n",
      "                                                                 \n",
      " depthwise_conv2d_3 (Depthwi  (None, 4, 4, 64)         640       \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_7 (Batc  (None, 4, 4, 64)         256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_7 (ReLU)              (None, 4, 4, 64)          0         \n",
      "                                                                 \n",
      " conv2d_4 (Conv2D)           (None, 4, 4, 128)         8320      \n",
      "                                                                 \n",
      " batch_normalization_8 (Batc  (None, 4, 4, 128)        512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_8 (ReLU)              (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " depthwise_conv2d_4 (Depthwi  (None, 4, 4, 128)        1280      \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_9 (Batc  (None, 4, 4, 128)        512       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " re_lu_9 (ReLU)              (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " conv2d_5 (Conv2D)           (None, 4, 4, 128)         16512     \n",
      "                                                                 \n",
      " batch_normalization_10 (Bat  (None, 4, 4, 128)        512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_10 (ReLU)             (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 2048)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               262272    \n",
      "                                                                 \n",
      " batch_normalization_11 (Bat  (None, 128)              512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_11 (ReLU)             (None, 128)               0         \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 10)                1290      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 301,082\n",
      "Trainable params: 299,418\n",
      "Non-trainable params: 1,664\n",
      "_________________________________________________________________\n",
      "Epoch 1/50\n",
      "157/157 [==============================] - 9s 49ms/step - loss: 2.5742 - accuracy: 0.1400 - val_loss: 2.6148 - val_accuracy: 0.1030\n",
      "Epoch 2/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 2.1040 - accuracy: 0.2286 - val_loss: 2.7812 - val_accuracy: 0.1000\n",
      "Epoch 3/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.9024 - accuracy: 0.2893 - val_loss: 2.6200 - val_accuracy: 0.1560\n",
      "Epoch 4/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.8071 - accuracy: 0.3235 - val_loss: 2.2137 - val_accuracy: 0.2190\n",
      "Epoch 5/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.7522 - accuracy: 0.3414 - val_loss: 1.8551 - val_accuracy: 0.3230\n",
      "Epoch 6/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.7003 - accuracy: 0.3655 - val_loss: 1.8514 - val_accuracy: 0.3140\n",
      "Epoch 7/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.6414 - accuracy: 0.3936 - val_loss: 1.7159 - val_accuracy: 0.3720\n",
      "Epoch 8/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.6064 - accuracy: 0.4108 - val_loss: 1.6513 - val_accuracy: 0.4000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.5662 - accuracy: 0.4227 - val_loss: 1.7367 - val_accuracy: 0.3720\n",
      "Epoch 10/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.5274 - accuracy: 0.4358 - val_loss: 1.6411 - val_accuracy: 0.4220\n",
      "Epoch 11/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.5034 - accuracy: 0.4414 - val_loss: 1.5357 - val_accuracy: 0.4600\n",
      "Epoch 12/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.4528 - accuracy: 0.4609 - val_loss: 1.4456 - val_accuracy: 0.4880\n",
      "Epoch 13/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.4314 - accuracy: 0.4721 - val_loss: 1.3785 - val_accuracy: 0.5090\n",
      "Epoch 14/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.3990 - accuracy: 0.4895 - val_loss: 1.3330 - val_accuracy: 0.5160\n",
      "Epoch 15/50\n",
      "157/157 [==============================] - 7s 45ms/step - loss: 1.3874 - accuracy: 0.4914 - val_loss: 1.4915 - val_accuracy: 0.4750\n",
      "Epoch 16/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.3590 - accuracy: 0.5008 - val_loss: 1.3274 - val_accuracy: 0.5190\n",
      "Epoch 17/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.3463 - accuracy: 0.5065 - val_loss: 1.3264 - val_accuracy: 0.5330\n",
      "Epoch 18/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.3186 - accuracy: 0.5184 - val_loss: 1.2465 - val_accuracy: 0.5600\n",
      "Epoch 19/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.2977 - accuracy: 0.5302 - val_loss: 1.2152 - val_accuracy: 0.5640\n",
      "Epoch 20/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.2796 - accuracy: 0.5347 - val_loss: 1.2760 - val_accuracy: 0.5570\n",
      "Epoch 21/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.2529 - accuracy: 0.5454 - val_loss: 1.2864 - val_accuracy: 0.5360\n",
      "Epoch 22/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.2343 - accuracy: 0.5546 - val_loss: 1.1768 - val_accuracy: 0.5830\n",
      "Epoch 23/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.2188 - accuracy: 0.5541 - val_loss: 1.1729 - val_accuracy: 0.5670\n",
      "Epoch 24/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.1940 - accuracy: 0.5730 - val_loss: 1.1531 - val_accuracy: 0.5910\n",
      "Epoch 25/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.1683 - accuracy: 0.5763 - val_loss: 1.2254 - val_accuracy: 0.5730\n",
      "Epoch 26/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.1657 - accuracy: 0.5817 - val_loss: 1.1307 - val_accuracy: 0.6080\n",
      "Epoch 27/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.1478 - accuracy: 0.5848 - val_loss: 1.0730 - val_accuracy: 0.6310\n",
      "Epoch 28/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.1364 - accuracy: 0.5912 - val_loss: 1.0865 - val_accuracy: 0.6140\n",
      "Epoch 29/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.1142 - accuracy: 0.6033 - val_loss: 1.0662 - val_accuracy: 0.6450\n",
      "Epoch 30/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.1082 - accuracy: 0.6079 - val_loss: 1.1342 - val_accuracy: 0.5950\n",
      "Epoch 31/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0951 - accuracy: 0.6054 - val_loss: 1.1631 - val_accuracy: 0.5890\n",
      "Epoch 32/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0851 - accuracy: 0.6092 - val_loss: 1.1317 - val_accuracy: 0.5960\n",
      "Epoch 33/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0695 - accuracy: 0.6120 - val_loss: 1.0899 - val_accuracy: 0.5980\n",
      "Epoch 34/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0620 - accuracy: 0.6142 - val_loss: 1.0739 - val_accuracy: 0.6060\n",
      "Epoch 35/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0335 - accuracy: 0.6254 - val_loss: 1.1089 - val_accuracy: 0.6180\n",
      "Epoch 36/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0243 - accuracy: 0.6330 - val_loss: 1.0206 - val_accuracy: 0.6460\n",
      "Epoch 37/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0288 - accuracy: 0.6302 - val_loss: 0.9872 - val_accuracy: 0.6600\n",
      "Epoch 38/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0095 - accuracy: 0.6403 - val_loss: 0.9958 - val_accuracy: 0.6420\n",
      "Epoch 39/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 1.0086 - accuracy: 0.6427 - val_loss: 1.0061 - val_accuracy: 0.6320\n",
      "Epoch 40/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9710 - accuracy: 0.6539 - val_loss: 1.0037 - val_accuracy: 0.6370\n",
      "Epoch 41/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9769 - accuracy: 0.6514 - val_loss: 0.9840 - val_accuracy: 0.6430\n",
      "Epoch 42/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9610 - accuracy: 0.6582 - val_loss: 0.9663 - val_accuracy: 0.6610\n",
      "Epoch 43/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9750 - accuracy: 0.6579 - val_loss: 1.0083 - val_accuracy: 0.6290\n",
      "Epoch 44/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9494 - accuracy: 0.6540 - val_loss: 1.0277 - val_accuracy: 0.6430\n",
      "Epoch 45/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9390 - accuracy: 0.6637 - val_loss: 1.0035 - val_accuracy: 0.6410\n",
      "Epoch 46/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9361 - accuracy: 0.6611 - val_loss: 0.9792 - val_accuracy: 0.6570\n",
      "Epoch 47/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9220 - accuracy: 0.6692 - val_loss: 0.9946 - val_accuracy: 0.6410\n",
      "Epoch 48/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9145 - accuracy: 0.6677 - val_loss: 0.9850 - val_accuracy: 0.6470\n",
      "Epoch 49/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9075 - accuracy: 0.6747 - val_loss: 0.9380 - val_accuracy: 0.6690\n",
      "Epoch 50/50\n",
      "157/157 [==============================] - 7s 46ms/step - loss: 0.9090 - accuracy: 0.6739 - val_loss: 0.9861 - val_accuracy: 0.6470\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: ./checkpoints/saved_model_v5/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: ./checkpoints/saved_model_v5/assets\n",
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpmh7q3mwk/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpmh7q3mwk/assets\n",
      "2022-11-13 16:56:41.693996: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
      "2022-11-13 16:56:41.694032: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n",
      "2022-11-13 16:56:41.694570: I tensorflow/cc/saved_model/reader.cc:45] Reading SavedModel from: /tmp/tmpmh7q3mwk\n",
      "2022-11-13 16:56:41.702559: I tensorflow/cc/saved_model/reader.cc:89] Reading meta graph with tags { serve }\n",
      "2022-11-13 16:56:41.702597: I tensorflow/cc/saved_model/reader.cc:130] Reading SavedModel debug info (if present) from: /tmp/tmpmh7q3mwk\n",
      "2022-11-13 16:56:41.723145: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:354] MLIR V1 optimization pass is not enabled\n",
      "2022-11-13 16:56:41.732670: I tensorflow/cc/saved_model/loader.cc:229] Restoring SavedModel bundle.\n",
      "2022-11-13 16:56:41.946794: I tensorflow/cc/saved_model/loader.cc:213] Running initialization op on SavedModel bundle at path: /tmp/tmpmh7q3mwk\n",
      "2022-11-13 16:56:42.001289: I tensorflow/cc/saved_model/loader.cc:305] SavedModel load for tags { serve }; Status: success: OK. Took 306718 microseconds.\n",
      "2022-11-13 16:56:42.118441: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:268] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n"
     ]
    }
   ],
   "source": [
    "# train all\n",
    "for MODEL_VERSION in range(5,6):\n",
    "    if MODEL_VERSION == 1:\n",
    "        model = model_v1()\n",
    "        model_name = \"v1\"\n",
    "    elif MODEL_VERSION == 2:\n",
    "        model = model_v2()\n",
    "        model_name = \"v2\"\n",
    "    elif MODEL_VERSION == 3:\n",
    "        model = model_v3()\n",
    "        model_name = \"v3\"\n",
    "    elif MODEL_VERSION == 4:\n",
    "        model = model_v4()\n",
    "        model_name = \"v4\"\n",
    "    elif MODEL_VERSION == 5:\n",
    "        model = model_v5()\n",
    "        model_name = \"v5\"\n",
    "        \n",
    "    model.summary()\n",
    "    checkpoint_path = f\"./checkpoints/saved_model_{model_name}/\"\n",
    "    model.compile(optimizer='adam', loss=keras.losses.categorical_crossentropy, metrics=['accuracy'])\n",
    "    history = model.fit(train_images, train_labels, batch_size=64, epochs=50, # Add more epochs to get better results\n",
    "                      validation_data=(test_images, test_labels))\n",
    "    model.save(checkpoint_path)\n",
    "    tflite_model_file = pathlib.Path(f\"cifar10_model_{model_name}_fp32.tflite\")\n",
    "    # Converting a tf.Keras model to a TensorFlow Lite model.\n",
    "    converter = tf.lite.TFLiteConverter.from_keras_model(model)\n",
    "    tflite_model = converter.convert()\n",
    "\n",
    "    # Save the unquantized/float model:\n",
    "    tflite_model_file.write_bytes(tflite_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Uk9v8V2gCFXF",
    "outputId": "2568277e-f3cb-4eec-ce03-6fc7ad5bd197"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d_6 (Conv2D)           (None, 32, 32, 16)        1216      \n",
      "                                                                 \n",
      " batch_normalization_12 (Bat  (None, 32, 32, 16)       64        \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_12 (ReLU)             (None, 32, 32, 16)        0         \n",
      "                                                                 \n",
      " depthwise_conv2d_5 (Depthwi  (None, 16, 16, 16)       160       \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_13 (Bat  (None, 16, 16, 16)       64        \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_13 (ReLU)             (None, 16, 16, 16)        0         \n",
      "                                                                 \n",
      " conv2d_7 (Conv2D)           (None, 16, 16, 32)        544       \n",
      "                                                                 \n",
      " batch_normalization_14 (Bat  (None, 16, 16, 32)       128       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_14 (ReLU)             (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " dropout_4 (Dropout)         (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " depthwise_conv2d_6 (Depthwi  (None, 16, 16, 32)       320       \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_15 (Bat  (None, 16, 16, 32)       128       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_15 (ReLU)             (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " conv2d_8 (Conv2D)           (None, 16, 16, 48)        1584      \n",
      "                                                                 \n",
      " batch_normalization_16 (Bat  (None, 16, 16, 48)       192       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_16 (ReLU)             (None, 16, 16, 48)        0         \n",
      "                                                                 \n",
      " depthwise_conv2d_7 (Depthwi  (None, 8, 8, 48)         480       \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_17 (Bat  (None, 8, 8, 48)         192       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_17 (ReLU)             (None, 8, 8, 48)          0         \n",
      "                                                                 \n",
      " conv2d_9 (Conv2D)           (None, 8, 8, 64)          3136      \n",
      "                                                                 \n",
      " batch_normalization_18 (Bat  (None, 8, 8, 64)         256       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_18 (ReLU)             (None, 8, 8, 64)          0         \n",
      "                                                                 \n",
      " dropout_5 (Dropout)         (None, 8, 8, 64)          0         \n",
      "                                                                 \n",
      " depthwise_conv2d_8 (Depthwi  (None, 4, 4, 64)         640       \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_19 (Bat  (None, 4, 4, 64)         256       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_19 (ReLU)             (None, 4, 4, 64)          0         \n",
      "                                                                 \n",
      " conv2d_10 (Conv2D)          (None, 4, 4, 128)         8320      \n",
      "                                                                 \n",
      " batch_normalization_20 (Bat  (None, 4, 4, 128)        512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_20 (ReLU)             (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " depthwise_conv2d_9 (Depthwi  (None, 4, 4, 128)        1280      \n",
      " seConv2D)                                                       \n",
      "                                                                 \n",
      " batch_normalization_21 (Bat  (None, 4, 4, 128)        512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_21 (ReLU)             (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " conv2d_11 (Conv2D)          (None, 4, 4, 128)         16512     \n",
      "                                                                 \n",
      " batch_normalization_22 (Bat  (None, 4, 4, 128)        512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_22 (ReLU)             (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " dropout_6 (Dropout)         (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " flatten_1 (Flatten)         (None, 2048)              0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 128)               262272    \n",
      "                                                                 \n",
      " batch_normalization_23 (Bat  (None, 128)              512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " re_lu_23 (ReLU)             (None, 128)               0         \n",
      "                                                                 \n",
      " dropout_7 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 10)                1290      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 301,082\n",
      "Trainable params: 299,418\n",
      "Non-trainable params: 1,664\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "MODEL_VERSION = 5\n",
    "\n",
    "if MODEL_VERSION == 1:\n",
    "    model = model_v1()\n",
    "    model_name = \"v1\"\n",
    "elif MODEL_VERSION == 2:\n",
    "    model = model_v2()\n",
    "    model_name = \"v2\"\n",
    "elif MODEL_VERSION == 3:\n",
    "    model = model_v3()\n",
    "    model_name = \"v3\"\n",
    "elif MODEL_VERSION == 4:\n",
    "    model = model_v4()\n",
    "    model_name = \"v4\"\n",
    "elif MODEL_VERSION == 5:\n",
    "    model = model_v5()\n",
    "    model_name = \"v5\"\n",
    "\n",
    "# Checking the model summary\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mgrXAkCkCRzL"
   },
   "source": [
    "# Compile and Train the model (or load the pretrained one)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "PZL5RClrCN1h"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'v5'}\n"
     ]
    }
   ],
   "source": [
    "#checkpoint_path = \"gdrive/MyDrive/cifar10/saved_model/my_model\"\n",
    "print ({model_name})\n",
    "checkpoint_path = f\"./checkpoints/saved_model_{model_name}/\"\n",
    "train_again = False\n",
    "\n",
    "if os.path.exists(checkpoint_path) and not train_again:\n",
    "    model = tf.keras.models.load_model(checkpoint_path)\n",
    "    history = None\n",
    "else:\n",
    "    model.compile(optimizer='adam', loss=keras.losses.categorical_crossentropy, metrics=['accuracy'])\n",
    "    history = model.fit(train_images, train_labels, batch_size=64, epochs=2, # Add more epochs to get better results\n",
    "                      validation_data=(test_images, test_labels))\n",
    "    model.save(checkpoint_path)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32/32 [==============================] - 0s 9ms/step\n",
      "Trained model Accuracy: 64.7%\n"
     ]
    }
   ],
   "source": [
    "# Making the Predictions\n",
    "pred = model.predict(test_images)\n",
    "accuracy = 100 * np.sum(np.argmax(pred, 1) == np.argmax(test_labels, 1)) / len(test_labels)\n",
    "print(f\"Trained model Accuracy: {accuracy}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JfryV1T7EhYG"
   },
   "source": [
    "# Visualize training results\n",
    "\n",
    "Only available if you have trained the model in this session"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "kxQKwWtEEjIP"
   },
   "outputs": [],
   "source": [
    "if history:\n",
    "    # Loss curve\n",
    "    plt.figure(figsize=[6,4])\n",
    "    plt.plot(history.history['loss'], 'black', linewidth=2.0)\n",
    "    plt.plot(history.history['val_loss'], 'green', linewidth=2.0)\n",
    "    plt.legend(['Training Loss', 'Validation Loss'], fontsize=14)\n",
    "    plt.xlabel('Epochs', fontsize=10)\n",
    "    plt.ylabel('Loss', fontsize=10)\n",
    "    plt.title('Loss Curves', fontsize=12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "7Ox9LUAfEqR8"
   },
   "outputs": [],
   "source": [
    "if history:\n",
    "    # Accuracy curve\n",
    "    plt.figure(figsize=[6,4])\n",
    "    plt.plot(history.history['accuracy'], 'black', linewidth=2.0)\n",
    "    plt.plot(history.history['val_accuracy'], 'blue', linewidth=2.0)\n",
    "    plt.legend(['Training Accuracy', 'Validation Accuracy'], fontsize=14)\n",
    "    plt.xlabel('Epochs', fontsize=10)\n",
    "    plt.ylabel('Accuracy', fontsize=10)\n",
    "    plt.title('Accuracy Curves', fontsize=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aDGTyZ-SEwQD"
   },
   "source": [
    "# See the model at work"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 830
    },
    "id": "UaWOvQi4EycM",
    "outputId": "7d2b2819-32e0-4d5d-a0ce-1a68cf2b31f1"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x1500 with 25 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting the Actual vs. Predicted results\n",
    "# Converting the predictions into label index \n",
    "pred_classes = np.argmax(pred, axis=1)\n",
    "\n",
    "fig, axes = plt.subplots(5, 5, figsize=(15,15))\n",
    "axes = axes.ravel()\n",
    "\n",
    "for i in np.arange(0, 25):\n",
    "    axes[i].imshow((128*(test_images[i]+1)).astype(np.uint8))\n",
    "    axes[i].set_title(f\"True: {class_names[np.argmax(test_labels[i])]}\\nPredict: {class_names[pred_classes[i]]}\", color=\"r\" if class_names[np.argmax(test_labels[i])]!=class_names[pred_classes[i]] else \"b\")\n",
    "    axes[i].axis('off')\n",
    "    plt.subplots_adjust(wspace=1)\n",
    "    img = Image.fromarray(np.uint8(128*(test_images[i]+1)))\n",
    "    img.save(f\"samples/cifar_test_{i}_{np.argmax(test_labels[i])}.ppm\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hHnm2AaEFFLE"
   },
   "source": [
    "# Convert to tflite"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "UhLXJ5mCFJ8W"
   },
   "outputs": [],
   "source": [
    "# Helper function to run inference on a TFLite model\n",
    "def test_tflite_model(tflite_file, test_images, test_labels):\n",
    "    # Initialize the interpreter\n",
    "    interpreter = tf.lite.Interpreter(model_path=str(tflite_file))\n",
    "    interpreter.allocate_tensors()\n",
    "\n",
    "    input_details = interpreter.get_input_details()[0]\n",
    "    output_details = interpreter.get_output_details()[0]\n",
    "\n",
    "    predictions = np.zeros((len(test_images),), dtype=int)\n",
    "    for i, (test_image, test_label) in enumerate(tqdm(zip(test_images, test_labels), total=len(test_labels))):\n",
    "        # Check if the input type is quantized, then rescale input data to uint8\n",
    "        if input_details['dtype'] == np.uint8:\n",
    "            input_scale, input_zero_point = input_details[\"quantization\"]\n",
    "            test_image = test_image / input_scale + input_zero_point\n",
    "\n",
    "        test_image = np.expand_dims(test_image, axis=0).astype(input_details[\"dtype\"])\n",
    "        interpreter.set_tensor(input_details[\"index\"], test_image)\n",
    "        interpreter.invoke()\n",
    "        output = interpreter.get_tensor(output_details[\"index\"])[0]\n",
    "\n",
    "        predictions[i] = output.argmax()\n",
    "\n",
    "    test_labels_not_one_hot = np.argmax(test_labels, 1)\n",
    "    accuracy = (np.sum(test_labels_not_one_hot == predictions) * 100) / len(test_images)\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "tflite_model_file = pathlib.Path(f\"cifar10_model_{model_name}_fp32.tflite\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "s_w0Bbs9FGjv",
    "outputId": "fcee918c-1e4b-45bd-e5ef-385c2ac6f226"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpwsvc0d0x/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpwsvc0d0x/assets\n",
      "2022-11-13 17:02:41.215386: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
      "2022-11-13 17:02:41.215413: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n",
      "2022-11-13 17:02:41.215559: I tensorflow/cc/saved_model/reader.cc:45] Reading SavedModel from: /tmp/tmpwsvc0d0x\n",
      "2022-11-13 17:02:41.222768: I tensorflow/cc/saved_model/reader.cc:89] Reading meta graph with tags { serve }\n",
      "2022-11-13 17:02:41.222807: I tensorflow/cc/saved_model/reader.cc:130] Reading SavedModel debug info (if present) from: /tmp/tmpwsvc0d0x\n",
      "2022-11-13 17:02:41.248152: I tensorflow/cc/saved_model/loader.cc:229] Restoring SavedModel bundle.\n",
      "2022-11-13 17:02:41.350965: I tensorflow/cc/saved_model/loader.cc:213] Running initialization op on SavedModel bundle at path: /tmp/tmpwsvc0d0x\n",
      "2022-11-13 17:02:41.396452: I tensorflow/cc/saved_model/loader.cc:305] SavedModel load for tags { serve }; Status: success: OK. Took 180893 microseconds.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "1202096"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Converting a tf.Keras model to a TensorFlow Lite model.\n",
    "converter = tf.lite.TFLiteConverter.from_keras_model(model)\n",
    "tflite_model = converter.convert()\n",
    "\n",
    "# Save the unquantized/float model:\n",
    "tflite_model_file.write_bytes(tflite_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kRlsyupjGFaa",
    "outputId": "0991c086-4d83-4088-a9cf-72bb90f37f26"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO: Created TensorFlow Lite XNNPACK delegate for CPU.\n",
      "100%|██████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 6167.91it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Float model accuracy: 64.7\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "fp32_accuracy = test_tflite_model(tflite_model_file, test_images, test_labels)\n",
    "print(f\"\\nFloat model accuracy: {fp32_accuracy}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-VqlDQv9HlJP"
   },
   "source": [
    "## Quantize to INT8 using PTQ\n",
    "\n",
    "We use post-training (integer) quantization (PTQ) to quantize the model (https://www.tensorflow.org/lite/performance/post_training_integer_quant).\n",
    "\n",
    "Weights are quantized directly from their values (they are constants), activations on the other hand depend on the input data. Hence we need to provide a calibration dataset to the quantizer so that it can run inference on it and collect the statistics of each layer in order to quantize the values in those ranges, i.e. Layer1 -> [-3.0, 6.0], Layer2 -> [-1.0, 2.5], ...\n",
    "\n",
    "As calibration dataset we need representative data of our use case. They cannot be the testing set, we are \"learning\" the statistics so using test dataset would be cheating. A subset of the training is tipycally used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "v4Vf9wK0GqZ0",
    "outputId": "e3c201f1-b413-4dea-e3e5-7c6bce8df55a"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 11). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpiqmsds1p/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpiqmsds1p/assets\n",
      "/home/cioflanc/miniconda3/envs/gapcifar10/lib/python3.9/site-packages/tensorflow/lite/python/convert.py:766: UserWarning: Statistics for quantized inputs were expected, but not specified; continuing anyway.\n",
      "  warnings.warn(\"Statistics for quantized inputs were expected, but not \"\n",
      "2022-11-13 17:02:51.682754: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
      "2022-11-13 17:02:51.682780: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n",
      "2022-11-13 17:02:51.682938: I tensorflow/cc/saved_model/reader.cc:45] Reading SavedModel from: /tmp/tmpiqmsds1p\n",
      "2022-11-13 17:02:51.694511: I tensorflow/cc/saved_model/reader.cc:89] Reading meta graph with tags { serve }\n",
      "2022-11-13 17:02:51.694548: I tensorflow/cc/saved_model/reader.cc:130] Reading SavedModel debug info (if present) from: /tmp/tmpiqmsds1p\n",
      "2022-11-13 17:02:51.725910: I tensorflow/cc/saved_model/loader.cc:229] Restoring SavedModel bundle.\n",
      "2022-11-13 17:02:51.832182: I tensorflow/cc/saved_model/loader.cc:213] Running initialization op on SavedModel bundle at path: /tmp/tmpiqmsds1p\n",
      "2022-11-13 17:02:51.878216: I tensorflow/cc/saved_model/loader.cc:305] SavedModel load for tags { serve }; Status: success: OK. Took 195278 microseconds.\n",
      "fully_quantize: 0, inference_type: 6, input_inference_type: UINT8, output_inference_type: UINT8\n"
     ]
    }
   ],
   "source": [
    "def representative_data_gen():\n",
    "  for input_value in tf.data.Dataset.from_tensor_slices(train_images).batch(1).take(100):\n",
    "    yield [input_value]\n",
    "\n",
    "converter = tf.lite.TFLiteConverter.from_keras_model(model)\n",
    "converter.optimizations = [tf.lite.Optimize.DEFAULT]\n",
    "converter.representative_dataset = representative_data_gen\n",
    "# Ensure that if any ops can't be quantized, the converter throws an error\n",
    "converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]\n",
    "# Set the input and output tensors to uint8 (APIs added in r2.3)\n",
    "converter.inference_input_type = tf.uint8\n",
    "converter.inference_output_type = tf.uint8\n",
    "\n",
    "tflite_model_quant = converter.convert()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Declare the saving path. Make sure that the directory exists.\n",
    "tflite_quant_model_file = pathlib.Path(f\"model/cifar10_{model_name}_uint8.tflite\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "329888"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Save the unquantized/float model:\n",
    "tflite_quant_model_file.write_bytes(tflite_model_quant)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Qm4RO-GIHCdW",
    "outputId": "6219c701-5212-463b-bb9f-8ed8839c0cd1"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 3171.64it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Quantized model accuracy: 64.3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "quant_accuracy = test_tflite_model(tflite_quant_model_file, test_images, test_labels)\n",
    "print(f\"Quantized model accuracy: {quant_accuracy}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Quantize to INT8 using QAT\n",
    "\n",
    "After performing post-training quantization, we might notice an accuracy drop between the FP32 and INT8 networks. As the weights are kept constant during PTQ and the activations' range is analysed only with respect to the input (calibration) data, such a quantization strategy, albeit quick, can render a model inefficient.\n",
    "\n",
    "Quantization-aware training (QAT) updates the model's weights whilst reducing the bitwidth. A comprehensive guide on TensorFlow-based QAT is available [here](https://www.tensorflow.org/model_optimization/guide/quantization/training_comprehensive_guide)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " quantize_layer (QuantizeLay  (None, 32, 32, 3)        3         \n",
      " er)                                                             \n",
      "                                                                 \n",
      " quant_conv2d (QuantizeWrapp  (None, 32, 32, 16)       1249      \n",
      " erV2)                                                           \n",
      "                                                                 \n",
      " quant_batch_normalization (  (None, 32, 32, 16)       65        \n",
      " QuantizeWrapperV2)                                              \n",
      "                                                                 \n",
      " quant_re_lu (QuantizeWrappe  (None, 32, 32, 16)       3         \n",
      " rV2)                                                            \n",
      "                                                                 \n",
      " quant_depthwise_conv2d (Qua  (None, 16, 16, 16)       163       \n",
      " ntizeWrapperV2)                                                 \n",
      "                                                                 \n",
      " quant_batch_normalization_1  (None, 16, 16, 16)       65        \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_1 (QuantizeWrap  (None, 16, 16, 16)       3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_conv2d_1 (QuantizeWra  (None, 16, 16, 32)       609       \n",
      " pperV2)                                                         \n",
      "                                                                 \n",
      " quant_batch_normalization_2  (None, 16, 16, 32)       129       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_2 (QuantizeWrap  (None, 16, 16, 32)       3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_dropout (QuantizeWrap  (None, 16, 16, 32)       1         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_depthwise_conv2d_1 (Q  (None, 16, 16, 32)       323       \n",
      " uantizeWrapperV2)                                               \n",
      "                                                                 \n",
      " quant_batch_normalization_3  (None, 16, 16, 32)       129       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_3 (QuantizeWrap  (None, 16, 16, 32)       3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_conv2d_2 (QuantizeWra  (None, 16, 16, 48)       1681      \n",
      " pperV2)                                                         \n",
      "                                                                 \n",
      " quant_batch_normalization_4  (None, 16, 16, 48)       193       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_4 (QuantizeWrap  (None, 16, 16, 48)       3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_depthwise_conv2d_2 (Q  (None, 8, 8, 48)         483       \n",
      " uantizeWrapperV2)                                               \n",
      "                                                                 \n",
      " quant_batch_normalization_5  (None, 8, 8, 48)         193       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_5 (QuantizeWrap  (None, 8, 8, 48)         3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_conv2d_3 (QuantizeWra  (None, 8, 8, 64)         3265      \n",
      " pperV2)                                                         \n",
      "                                                                 \n",
      " quant_batch_normalization_6  (None, 8, 8, 64)         257       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_6 (QuantizeWrap  (None, 8, 8, 64)         3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_dropout_1 (QuantizeWr  (None, 8, 8, 64)         1         \n",
      " apperV2)                                                        \n",
      "                                                                 \n",
      " quant_depthwise_conv2d_3 (Q  (None, 4, 4, 64)         643       \n",
      " uantizeWrapperV2)                                               \n",
      "                                                                 \n",
      " quant_batch_normalization_7  (None, 4, 4, 64)         257       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_7 (QuantizeWrap  (None, 4, 4, 64)         3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_conv2d_4 (QuantizeWra  (None, 4, 4, 128)        8577      \n",
      " pperV2)                                                         \n",
      "                                                                 \n",
      " quant_batch_normalization_8  (None, 4, 4, 128)        513       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_8 (QuantizeWrap  (None, 4, 4, 128)        3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_depthwise_conv2d_4 (Q  (None, 4, 4, 128)        1283      \n",
      " uantizeWrapperV2)                                               \n",
      "                                                                 \n",
      " quant_batch_normalization_9  (None, 4, 4, 128)        513       \n",
      "  (QuantizeWrapperV2)                                            \n",
      "                                                                 \n",
      " quant_re_lu_9 (QuantizeWrap  (None, 4, 4, 128)        3         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_conv2d_5 (QuantizeWra  (None, 4, 4, 128)        16769     \n",
      " pperV2)                                                         \n",
      "                                                                 \n",
      " quant_batch_normalization_1  (None, 4, 4, 128)        513       \n",
      " 0 (QuantizeWrapperV2)                                           \n",
      "                                                                 \n",
      " quant_re_lu_10 (QuantizeWra  (None, 4, 4, 128)        3         \n",
      " pperV2)                                                         \n",
      "                                                                 \n",
      " quant_dropout_2 (QuantizeWr  (None, 4, 4, 128)        1         \n",
      " apperV2)                                                        \n",
      "                                                                 \n",
      " quant_flatten (QuantizeWrap  (None, 2048)             1         \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      " quant_dense (QuantizeWrappe  (None, 128)              262275    \n",
      " rV2)                                                            \n",
      "                                                                 \n",
      " quant_batch_normalization_1  (None, 128)              515       \n",
      " 1 (QuantizeWrapperV2)                                           \n",
      "                                                                 \n",
      " quant_re_lu_11 (QuantizeWra  (None, 128)              3         \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " pperV2)                                                         \n",
      "                                                                 \n",
      " quant_dropout_3 (QuantizeWr  (None, 128)              1         \n",
      " apperV2)                                                        \n",
      "                                                                 \n",
      " quant_dense_1 (QuantizeWrap  (None, 10)               1295      \n",
      " perV2)                                                          \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 302,001\n",
      "Trainable params: 299,418\n",
      "Non-trainable params: 2,583\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Clone the trained TensorFlow model\n",
    "\n",
    "quantize_model = tfmot.quantization.keras.quantize_model\n",
    "tf_model_qat = quantize_model(model)\n",
    "\n",
    "# We need to compile again the model\n",
    "tf_model_qat.compile(optimizer='adam', loss=keras.losses.categorical_crossentropy, metrics=['accuracy'])\n",
    "\n",
    "tf_model_qat.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "157/157 [==============================] - 9s 60ms/step - loss: 0.8138 - accuracy: 0.7113 - val_loss: 0.9073 - val_accuracy: 0.6830\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as conv2d_layer_call_fn, conv2d_layer_call_and_return_conditional_losses, _jit_compiled_convolution_op, re_lu_layer_call_fn, re_lu_layer_call_and_return_conditional_losses while saving (showing 5 of 71). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: ./checkpoints/saved_model_qat_v5/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: ./checkpoints/saved_model_qat_v5/assets\n"
     ]
    }
   ],
   "source": [
    "# Finetune the fake-quantized model on a subset of the training data\n",
    "\n",
    "checkpoint_path = f\"./checkpoints/saved_model_qat_{model_name}/\"\n",
    "\n",
    "history = tf_model_qat.fit(train_images, train_labels, batch_size=64, epochs=1, # Add more epochs to get better results\n",
    "                  validation_data=(test_images, test_labels))\n",
    "tf_model_qat.save(checkpoint_path)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32/32 [==============================] - 0s 11ms/step\n",
      "Trained model Accuracy: 68.3%\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the QAT model\n",
    "\n",
    "pred = tf_model_qat.predict(test_images)\n",
    "accuracy = 100 * np.sum(np.argmax(pred, 1) == np.argmax(test_labels, 1)) / len(test_labels)\n",
    "print(f\"Trained model Accuracy: {accuracy}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as conv2d_layer_call_fn, conv2d_layer_call_and_return_conditional_losses, _jit_compiled_convolution_op, re_lu_layer_call_fn, re_lu_layer_call_and_return_conditional_losses while saving (showing 5 of 71). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpj11sofnn/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpj11sofnn/assets\n",
      "/home/cioflanc/miniconda3/envs/gapcifar10/lib/python3.9/site-packages/tensorflow/lite/python/convert.py:766: UserWarning: Statistics for quantized inputs were expected, but not specified; continuing anyway.\n",
      "  warnings.warn(\"Statistics for quantized inputs were expected, but not \"\n",
      "2022-11-13 17:07:26.167136: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
      "2022-11-13 17:07:26.167168: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n",
      "2022-11-13 17:07:26.167321: I tensorflow/cc/saved_model/reader.cc:45] Reading SavedModel from: /tmp/tmpj11sofnn\n",
      "2022-11-13 17:07:26.192685: I tensorflow/cc/saved_model/reader.cc:89] Reading meta graph with tags { serve }\n",
      "2022-11-13 17:07:26.192725: I tensorflow/cc/saved_model/reader.cc:130] Reading SavedModel debug info (if present) from: /tmp/tmpj11sofnn\n",
      "2022-11-13 17:07:26.265754: I tensorflow/cc/saved_model/loader.cc:229] Restoring SavedModel bundle.\n",
      "2022-11-13 17:07:26.624067: I tensorflow/cc/saved_model/loader.cc:213] Running initialization op on SavedModel bundle at path: /tmp/tmpj11sofnn\n",
      "2022-11-13 17:07:26.719890: I tensorflow/cc/saved_model/loader.cc:305] SavedModel load for tags { serve }; Status: success: OK. Took 552568 microseconds.\n",
      "fully_quantize: 0, inference_type: 6, input_inference_type: UINT8, output_inference_type: UINT8\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "331888"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quantize and save the model\n",
    "\n",
    "def representative_data_gen():\n",
    "  for input_value in tf.data.Dataset.from_tensor_slices(train_images).batch(1).take(100):\n",
    "    yield [input_value]\n",
    "\n",
    "converter_qat = tf.lite.TFLiteConverter.from_keras_model(tf_model_qat)\n",
    "converter_qat.optimizations = [tf.lite.Optimize.DEFAULT]\n",
    "converter_qat.representative_dataset = representative_data_gen\n",
    "# Ensure that if any ops can't be quantized, the converter throws an error\n",
    "converter_qat.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]\n",
    "# Set the input and output tensors to uint8 (APIs added in r2.3)\n",
    "converter_qat.inference_input_type = tf.uint8\n",
    "converter_qat.inference_output_type = tf.uint8\n",
    "\n",
    "tflite_model_qat_quant = converter_qat.convert()\n",
    "\n",
    "tflite_quant_qat_model_file = pathlib.Path(f\"model/cifar10_qat_{model_name}_uint8.tflite\")\n",
    "tflite_quant_qat_model_file.write_bytes(tflite_model_qat_quant)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 3107.15it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Quantized model accuracy: 68.6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the QAT model\n",
    "quant_qat_accuracy = test_tflite_model(tflite_quant_qat_model_file, test_images, test_labels)\n",
    "print(f\"Quantized model accuracy: {quant_qat_accuracy}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Did the accuracy increase compared to the PTQ solution? By how much?\n",
    "<br>\n",
    "Is this due to the benefits of QAT? How can you tell?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# NNTool Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def nntool_inference(graph, test_images, test_labels):\n",
    "    predictions = np.zeros((len(test_images),), dtype=int)\n",
    "    for i, (test_image, test_label) in enumerate(tqdm(zip(test_images, test_labels), total=len(test_labels))):\n",
    "        output = graph.execute([test_image], dequantize=True)\n",
    "        predictions[i] = output[-1][0].argmax()\n",
    "\n",
    "    test_labels_not_one_hot = np.argmax(test_labels, 1)\n",
    "    accuracy = (np.sum(test_labels_not_one_hot == predictions) * 100) / len(test_images)\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "TFLITE_PATH = \"model/cifar10_model_uint8.tflite\"\n",
    "NE16 = False\n",
    "G = NNGraph.load_graph(TFLITE_PATH, load_quantization=True)\n",
    "G.quantize(None, graph_options={\"use_ne16\": NE16, \"hwc\": True})\n",
    "G.adjust_order()\n",
    "G.fusions(\"scaled_match_group\")\n",
    "#G.draw(quant_labels=True, fusions=True, view=False, filepath=\"nntool_graph\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nntool_quant_accuracy = nntool_inference(G, test_images[:1000], test_labels[:1000])\n",
    "print(f\"Quantized model accuracy: {nntool_quant_accuracy}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "interpreter = tf.lite.Interpreter(model_path=str(TFLITE_PATH))\n",
    "interpreter.allocate_tensors()\n",
    "input_details = interpreter.get_input_details()[0]\n",
    "output_details = interpreter.get_output_details()[0]\n",
    "\n",
    "test_image = test_images[0]\n",
    "# Check if the input type is quantized, then rescale input data to uint8\n",
    "if input_details['dtype'] == np.uint8:\n",
    "    input_scale, input_zero_point = input_details[\"quantization\"]\n",
    "    test_image = test_image / input_scale + input_zero_point\n",
    "test_image = np.expand_dims(test_images[0], axis=0).astype(input_details[\"dtype\"])\n",
    "\n",
    "interpreter.set_tensor(input_details[\"index\"], test_image)\n",
    "interpreter.invoke()\n",
    "output_scale, output_zero_point = output_details[\"quantization\"]\n",
    "tflite_output = (interpreter.get_tensor(output_details[\"index\"])[0] - output_zero_point) * output_scale\n",
    "\n",
    "nntool_output = G.execute([test_images[0]], dequantize=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.plot(nntool_output[-1][0])\n",
    "plt.plot(tflite_output)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Quantize in NNTool"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "TFLITE_PATH = \"model/cifar10_model_uint8.tflite\"\n",
    "NE16 = True\n",
    "G = NNGraph.load_graph(TFLITE_PATH, load_quantization=True)\n",
    "\n",
    "stats = G.collect_statistics([np.array(data) for data in tf.data.Dataset.from_tensor_slices(train_images).batch(1).take(100)])\n",
    "G.quantize(stats, graph_options={\"use_ne16\": NE16, \"hwc\": True})\n",
    "G.adjust_order()\n",
    "G.fusions(\"scaled_match_group\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nntool_quant_accuracy = nntool_inference(G, test_images[:1000], test_labels[:1000])\n",
    "print(f\"Quantized model accuracy: {nntool_quant_accuracy}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(G.qshow())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "CIFAR10_TFLITE.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}