examples/notebooks/2_Build_a_CNN_with_NETS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Build a CNN on MNIST Dataset**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 0. Set Up\n",
    "\n",
    "For this tutorial, we will be using ``numpy`` to generate data and all operations, ``matplotlib`` to render and plot images and of course ``nets`` for ``Tensor`` operations and computational graph creation. We will set the ``SEED`` for random numbers from ``numpy``, so each time we reload the notebook the results will be the same as numbers will be generated in the same way. Finally, we set an autoreload with ``%load_ext autoreload`` and ``%autoreload 2`` for debug mode purposes. This enable the notebook to reload the modules each time we run a cell, so it keeps everything up-to-date. More info [on StackOverflow](http://stackoverflow.com/questions/1907993/autoreload-of-modules-in-ipython).\n",
    "\n",
    "Let's load the packages we need to run our tests:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import random as rd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# This is a bit of magic to make matplotlib figures appear inline in the notebook\n",
    "# rather than in a new window.\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (10.0, 8.0) # set default size of plots\n",
    "plt.rcParams['image.interpolation'] = 'nearest'\n",
    "plt.style.use('seaborn-darkgrid')\n",
    "# Change the font ?\n",
    "font = {'size'   : 15}\n",
    "plt.rc('font', **font)\n",
    "\n",
    "# Some more magic so that the notebook will reload external python modules;\n",
    "# see http://stackoverflow.com/questions/1907993/autoreload-of-modules-in-ipython\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'nvidia-smi' n'est pas reconnu en tant que commande interne\n",
      "ou externe, un programme ex‚cutable ou un fichier de commandes.\n",
      "Your runtime has 8.5 gigabytes of available RAM\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from psutil import virtual_memory\n",
    "\n",
    "\n",
    "gpu_info = !nvidia-smi\n",
    "gpu_info = '\\n'.join(gpu_info)\n",
    "print(gpu_info)\n",
    "\n",
    "ram_gb = virtual_memory().total / 1e9\n",
    "print('Your runtime has {:.1f} gigabytes of available RAM\\n'.format(ram_gb))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "\n",
    "sys.path.append(\"../../\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's fix the seed\n",
    "\n",
    "SEED = 42\n",
    "rd.seed(SEED)\n",
    "np.random.seed(SEED)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Download the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Download train-images-idx3-ubyte.gz from http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to .data\\mnist\\train-images-idx3-ubyte.gz\n",
      "Downloading... 100% | [==================================================] | Done !\n",
      ".data\\mnist\\mnist-data-py\n",
      "Extracting... | NOTE: gzip files are not extracted, and moved to .data\\mnist\\mnist-data-py | Done !\n",
      "Download train-labels-idx1-ubyte.gz from http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to .data\\mnist\\train-labels-idx1-ubyte.gz\n",
      "Downloading... 100% | [==================================================] | Done !\n",
      ".data\\mnist\\mnist-data-py\n",
      "Extracting... | NOTE: gzip files are not extracted, and moved to .data\\mnist\\mnist-data-py | Done !\n",
      "Download t10k-labels-idx1-ubyte.gz from http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to .data\\mnist\\t10k-labels-idx1-ubyte.gz\n",
      "Downloading... 100% | [==================================================] | Done !\n",
      ".data\\mnist\\mnist-data-py\n",
      "Extracting... | NOTE: gzip files are not extracted, and moved to .data\\mnist\\mnist-data-py | Done !\n",
      "Download t10k-images-idx3-ubyte.gz from http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to .data\\mnist\\t10k-images-idx3-ubyte.gz\n",
      "Downloading... 100% | [==================================================] | Done !\n",
      ".data\\mnist\\mnist-data-py\n",
      "Extracting... | NOTE: gzip files are not extracted, and moved to .data\\mnist\\mnist-data-py | Done !\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dataset MNIST(\n",
       "   (data): Tensor(shape=(60000, 1, 28, 28), dtype=float64)\n",
       "   (labels): Tensor(shape=(60000,), dtype=uint8)\n",
       ")"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import random as rd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import nets\n",
    "import nets.datasets as datasets\n",
    "\n",
    "# IMPORTANT: normalize the data !\n",
    "# 1. reshape from (data_length, 32, 32, 3) to (data_length, 32 * 32 * 3)\n",
    "reshape = lambda x: x.reshape(-1, 784)\n",
    "# 2. scale the data: data = (data - mean) / std\n",
    "normalize = lambda x: (x - np.mean(x, axis=1, keepdims=True)) / np.std(x, axis=1, keepdims=True)\n",
    "# 3. all together\n",
    "transform = lambda x: normalize(reshape(x)).reshape(-1, 1, 28, 28)#.transpose(0, 3, 1, 2)\n",
    "\n",
    "# Download the training and testing data\n",
    "train_data, test_data = datasets.MNIST.splits(transform = transform)\n",
    "\n",
    "\n",
    "train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data.data.cuda()\n",
    "train_data.labels.cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 32 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# CIFAR10 classes\n",
    "# classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')\n",
    "\n",
    "# Configure the mosaic to display images\n",
    "fig = plt.figure(figsize = (20, 10))\n",
    "columns = 8\n",
    "rows = 4\n",
    "\n",
    "# Create the mosaic\n",
    "for i in range(1, columns * rows + 1):\n",
    "    # Choose randomly an image\n",
    "    image_index = rd.randint(0, len(train_data))\n",
    "    image, label = train_data[image_index]\n",
    "    # Plot it\n",
    "    fig.add_subplot(rows, columns, i)\n",
    "    plt.imshow(image.numpy().reshape(28, 28))\n",
    "    # Remove axis and display image's label\n",
    "    # plt.title(classes[label.data])\n",
    "    plt.axis('off')\n",
    "    \n",
    "# Render\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Load the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nets.data import Iterator\n",
    "\n",
    "BATCH_SIZE = 64\n",
    "\n",
    "iterator = Iterator(train_data, \n",
    "                    batch_size = BATCH_SIZE, \n",
    "                    shuffle = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Classifier(\n",
       "   (conv): Conv2d(in_channels=1, out_channels=1, filter_size=(3, 3), stride=1, pad=0, bias=True)\n",
       "   (pool): MaxPool2d(pool_size=(2, 2), stride=2, pad=0)\n",
       "   (layer1): Linear(input_dim=169, output_dim=100, bias=True)\n",
       "   (layer2): Linear(input_dim=100, output_dim=100, bias=True)\n",
       "   (layer3): Linear(input_dim=100, output_dim=10, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nets.nn as nn\n",
    "\n",
    "class Classifier(nn.Module):\n",
    "\n",
    "    def __init__(self, input_dim, hidden_dim, output_dim):\n",
    "        super().__init__()\n",
    "        \n",
    "        # Input batch shape: [batch_size, num_channel, height, width]\n",
    "        #         for MNIST: [batch_size,       1    ,   28  ,  28  ]\n",
    "        self.conv = nn.Conv2d(1, 1, (3, 3), stride=1, pad=0)\n",
    "        # Output shape: [batch_size, num_channel, height - 2, width - 2]\n",
    "        #           ie: [batch_size,        1   ,      26   ,     26   ]\n",
    "        \n",
    "        self.pool = nn.MaxPool2d((2, 2), stride=2, pad=0)\n",
    "        # Output shape: [batch_size, num_channel, height / 2, width / 2]\n",
    "        #           ie: [batch_size,        1   ,      13   ,     13   ]\n",
    "\n",
    "        \n",
    "        # Input shape: [batch_size, num_channel * height * width]\n",
    "        #          ie: [batch_size,              169            ]\n",
    "        self.layer1 = nn.Linear(input_dim, hidden_dim)\n",
    "        # Output shape: [batch_size, hidden_dim]\n",
    "        \n",
    "        self.layer2 = nn.Linear(hidden_dim, hidden_dim)\n",
    "        \n",
    "        self.layer3 = nn.Linear(hidden_dim, output_dim)\n",
    "        # Output shape: [batch_size, output_dim]\n",
    "\n",
    "    def forward(self, inputs):\n",
    "        # Convolution layer\n",
    "        out = self.conv(inputs)\n",
    "        # Decrease the dimensions\n",
    "        out = self.pool(out)\n",
    "        \n",
    "        # Transition from 4-d tensor to 2-d tensor\n",
    "        out = out.reshape(out.shape[0], -1)\n",
    "        # From: [batch_size, 1, H, W]\n",
    "        #   to: [batch_size, input_dim]\n",
    "        \n",
    "        # Classification layer(s)\n",
    "        out = nets.relu(self.layer1(out))\n",
    "        out = nets.relu(self.layer2(out))\n",
    "        \n",
    "        # Prediction layer\n",
    "        out = self.layer3(out)\n",
    "        return out\n",
    "\n",
    "\n",
    "model = Classifier(169, 100, 10)\n",
    "\n",
    "# Let's check the architecture\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.cuda()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.3. Criterion & Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nets.optim import SGD\n",
    "from nets.nn import CrossEntropyLoss, MSELoss\n",
    "from nets.nn.utils import one_hot\n",
    "\n",
    "# How much do we update the parameters\n",
    "LEARNING_RATE = 0.1\n",
    "\n",
    "optimizer = SGD(model.parameters(), lr=LEARNING_RATE)\n",
    "criterion = CrossEntropyLoss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion.cuda()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.4. Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch    0/10: : 938it [00:25, 36.64it/s, train loss=0.006496]                                                         \n",
      "  0%|                                                                                          | 0/937 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch:    0/10 | loss: 1.163E-02\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch    1/10: : 938it [00:48, 19.31it/s, train loss=0.003954]                                                         \n",
      "  0%|                                                                                          | 0/937 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch:    1/10 | loss: 6.256E-03\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch    2/10: : 938it [01:05, 14.36it/s, train loss=0.002692]                                                         \n",
      "  0%|                                                                                          | 0/937 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch:    2/10 | loss: 5.452E-03\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch    3/10: : 938it [01:22, 11.34it/s, train loss=0.002167]                                                         \n",
      "  0%|                                                                                          | 0/937 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch:    3/10 | loss: 4.811E-03\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch    4/10: : 938it [01:42,  9.14it/s, train loss=0.001924]                                                         \n",
      "  0%|                                                                                          | 0/937 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch:    4/10 | loss: 4.271E-03\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch    5/10:   2%|#1                                           | 23/937 [00:06<01:49,  8.34it/s, train loss=0.003703]"
     ]
    }
   ],
   "source": [
    "import time\n",
    "from tqdm import tqdm\n",
    "\n",
    "# Training iterations\n",
    "EPOCHS = 10\n",
    "\n",
    "#Display the time during training\n",
    "start = time.time()\n",
    "# Record the loss\n",
    "history = []\n",
    "\n",
    "# Run the simulation EPOCHS times\n",
    "for epoch in range(EPOCHS):\n",
    "    \n",
    "    # Keep track of the loss at each epoch\n",
    "    epoch_loss = 0.0\n",
    "    epoch_start = time.time()\n",
    "    \n",
    "    # Number of batches\n",
    "    N  = len(iterator)\n",
    "\n",
    "    # Predict for all batches\n",
    "    trange = tqdm(iterator, ascii=True, position=0, leave=True, total=len(iterator))\n",
    "    for idx, batch in enumerate(trange):\n",
    "        trange.set_description(f\"Epoch {epoch:4d}/{EPOCHS}\")\n",
    "\n",
    "        # Gradients accumulates, therefore we need to set them to zero at each iteration\n",
    "        model.zero_grad()\n",
    "        \n",
    "        # Get the data from the batch\n",
    "        image, label = batch    \n",
    "        # Run the forward pass on the model\n",
    "        predictions = model(image)\n",
    "        # Get how far are the predictions from the truth (aka gold labels)\n",
    "        label = one_hot(label, 10).astype(int)\n",
    "        loss = criterion(predictions, label)\n",
    "        \n",
    "        # Compute the gradient\n",
    "        loss.backward()\n",
    "        # Update the parameters\n",
    "        optimizer.step()\n",
    "        \n",
    "        # Record the loss for plotting\n",
    "        loss_avg = loss.item() / BATCH_SIZE # .item() is really important, it will free the memory from the computational graph\n",
    "        epoch_loss += loss_avg\n",
    "        # Let's plot a progress bar in the console\n",
    "        trange.set_postfix({f\"train loss\": f\"{loss_avg:.6f}\"})\n",
    "    \n",
    "    # Update the history of all previous loss\n",
    "    history.append(epoch_loss / N)\n",
    "    \n",
    "    # Check the mean loss for this loop\n",
    "    print(f\"\\nEpoch: {epoch:4d}/{EPOCHS} | loss: {epoch_loss / N:1.3E}\")\n",
    "\n",
    "# Duration of the total training loop\n",
    "print('\\ntime:', time.time() - start)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "Epoch:    0/10 | loss: 1.163E-02\n",
    "Epoch:    1/10 | loss: 6.256E-03\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(history)\n",
    "plt.title(\"Loss on MNIST dataset\")\n",
    "plt.xlabel(\"epochs\")\n",
    "plt.ylabel(\"loss\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Configure the mosaic to display images\n",
    "fig = plt.figure(figsize = (20, 10))\n",
    "columns = 4\n",
    "rows = 1\n",
    "\n",
    "# Conv2d filter\n",
    "fig.add_subplot(rows, columns, 1)\n",
    "conv_filter = model._modules['conv'].weight\n",
    "plt.imshow(conv_filter.numpy().reshape(3, 3))\n",
    "plt.title('Conv2d filter')\n",
    "plt.axis('off')\n",
    "\n",
    "# Linear1 weight\n",
    "fig.add_subplot(rows, columns, 2)\n",
    "layer1_weightr = model._modules['layer1'].weight\n",
    "plt.imshow(layer1_weightr.numpy())\n",
    "plt.title('Layer1 weight')\n",
    "plt.axis('off')\n",
    "  \n",
    "# Linear2 weight\n",
    "fig.add_subplot(rows, columns, 3)\n",
    "layer1_weightr = model._modules['layer2'].weight\n",
    "plt.imshow(layer1_weightr.numpy())\n",
    "plt.title('Layer2 weight')\n",
    "plt.axis('off')\n",
    "\n",
    "# Linear3 weight\n",
    "fig.add_subplot(rows, columns, 4)\n",
    "layer1_weightr = model._modules['layer3'].weight\n",
    "plt.imshow(layer1_weightr.numpy())\n",
    "plt.title('Layer3 weight')\n",
    "plt.axis('off')\n",
    "    \n",
    "# Render\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}