LeNet.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "MJejm-ZRMB0x"
   },
   "source": [
    "# LeNet-5\n",
    "In this homework you will use PyTorch to build and train a convolutional network. You are asked to implement the a modified LeNet-5 for MINIST digits classification. Different from the original LeNet-5, in this notebook we use max instead of average pooling, and use fully connected instead of gaussian final layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4rPoZrLgMB0y"
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torch.nn import functional\n",
    "import torchvision\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import time\n",
    "from IPython import display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "2YIzSTkMMB02"
   },
   "outputs": [],
   "source": [
    "# grayscale and inline plotting\n",
    "%matplotlib inline\n",
    "plt.rcParams['image.cmap'] = 'gray'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "HI2cHk8zufts"
   },
   "source": [
    "## Visualization tools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "FCkbK86VMB07"
   },
   "outputs": [],
   "source": [
    "def plot_image(image):\n",
    "    nr, nc = image.shape\n",
    "    extent = [-0.5, nc - 0.5, nr - 0.5, -0.5]\n",
    "    plt.imshow(image, extent=extent, origin='upper', interpolation='nearest')\n",
    "\n",
    "def visualize(t, loss, errcl, x5, x4, x3, x0, w1):\n",
    "\n",
    "    loss_avg = np.divide(\n",
    "        np.cumsum(loss[: t + 1]),\n",
    "        range(1, t + 2)\n",
    "    )\n",
    "\n",
    "    errcl_avg = np.divide(\n",
    "        np.cumsum(errcl[: t + 1]),\n",
    "        range(1, t + 2)\n",
    "    )\n",
    "\n",
    "    n_last_batches = np.min([20, t])\n",
    "    k = np.ones(n_last_batches * 2 + 1) / (n_last_batches + 1)\n",
    "    k[:n_last_batches] = 0\n",
    "    \n",
    "    errcl_sm = np.convolve(np.pad(errcl, mode=\"edge\", pad_width=n_last_batches), k, mode=\"valid\")\n",
    "    errcl_sm = errcl_sm[: len(errcl_avg)]\n",
    "\n",
    "    loss_sm = np.convolve(np.pad(loss, mode=\"edge\", pad_width=n_last_batches), k, mode=\"valid\")\n",
    "    loss_sm = loss_sm[: len(loss)]\n",
    "\n",
    "    display.clear_output(wait=True)\n",
    "\n",
    "    plt.subplot(4, 3, 1)\n",
    "    plt.plot(loss, label=\"loss\")\n",
    "    plt.plot(loss_sm, label=\"smothed loss\")\n",
    "    plt.plot(loss_avg, label=\"avg loss\")\n",
    "    plt.legend()\n",
    "    plt.ylim(0, np.max(loss)*1.05)\n",
    "    plt.title(\"loss: avg - %.4f, smoothed - %.4f, current - %.4f\"  % (loss_avg[t], loss_sm[t], loss[t]))\n",
    "\n",
    "    plt.subplot(4, 3, 2)\n",
    "    plt.plot(errcl, label=\"cl err\")\n",
    "    plt.plot(errcl_sm, label=\"smothed cl err\")\n",
    "    plt.plot(errcl_avg, label=\"avg cl err\")\n",
    "    plt.legend()\n",
    "    plt.ylim(0, np.max(errcl)*1.05)\n",
    "    plt.title(\"cl error: avg - %.4f, smoothed - %.4f, current - %.4f\"  % (errcl_avg[t], errcl_sm[t], errcl[t]))\n",
    "    \n",
    "    plt.subplot(4, 3, 3)\n",
    "    plt.bar(range(len(x5)), x5)\n",
    "    plt.title(\"class confidences\")\n",
    "\n",
    "    plt.subplot(4,3,4)\n",
    "    plt.hist(x4)\n",
    "    plt.title(\"F6 activations\")\n",
    "\n",
    "    plt.subplot(4,3,5)\n",
    "    plt.hist(x3)\n",
    "    plt.title(\"C5 activations\")\n",
    "\n",
    "    plt.subplot(4,3,6)\n",
    "    plot_image(x0)\n",
    "    plt.title(\"input image\")\n",
    "\n",
    "    for i in range(w1.shape[0]):\n",
    "        plt.subplot(4,3,7+i)\n",
    "        plot_image(w1[i])\n",
    "        plt.title(\"C1 kernel channel \" + str(i))\n",
    "\n",
    "    plt.subplots_adjust(wspace=0.5)\n",
    "    plt.subplots_adjust(hspace=0.5)\n",
    "    plt.gcf().set_size_inches(18.5, 10.5)\n",
    "    display.display(plt.gcf())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "cWBhNjnKul-6"
   },
   "source": [
    "## Data Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "gc7hIpVjuZG3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to data\\MNIST\\raw\\train-images-idx3-ubyte.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5823275359a242b388fea33c3c238042",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data\\MNIST\\raw\\train-images-idx3-ubyte.gz to data\\MNIST\\raw\n",
      "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to data\\MNIST\\raw\\train-labels-idx1-ubyte.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1a873b08d794f508134477e910335e3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data\\MNIST\\raw\\train-labels-idx1-ubyte.gz to data\\MNIST\\raw\n",
      "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to data\\MNIST\\raw\\t10k-images-idx3-ubyte.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eccaf7d1d891420daf86fd35beb66833",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data\\MNIST\\raw\\t10k-images-idx3-ubyte.gz to data\\MNIST\\raw\n",
      "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to data\\MNIST\\raw\\t10k-labels-idx1-ubyte.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b912d3f7921a4220824489fad5df7189",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting data\\MNIST\\raw\\t10k-labels-idx1-ubyte.gz to data\\MNIST\\raw\n",
      "Processing...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "..\\torch\\csrc\\utils\\tensor_numpy.cpp:141: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done!\n",
      "Training Data shape is:  [60000, 28, 28]\n",
      "Training Target shape is:  [60000]\n"
     ]
    }
   ],
   "source": [
    "# get data\n",
    "\n",
    "import urllib\n",
    "try:\n",
    "    # For python 2\n",
    "    class AppURLopener(urllib.FancyURLopener):\n",
    "        version = \"Mozilla/5.0\"\n",
    "\n",
    "    urllib._urlopener = AppURLopener()\n",
    "except AttributeError:\n",
    "    # For python 3\n",
    "    opener = urllib.request.build_opener()\n",
    "    opener.addheaders = [('User-agent', 'Mozilla/5.0')]\n",
    "    urllib.request.install_opener(opener)\n",
    "\n",
    "mnist = torchvision.datasets.MNIST(root='data', train=True, download=True) # train data only\n",
    "trainimages = mnist.data\n",
    "trainlabels = mnist.targets\n",
    "\n",
    "# check training data shape\n",
    "print (\"Training Data shape is: \", list(trainimages.size()))\n",
    "print (\"Training Target shape is: \", list(trainlabels.size()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "NfsxXO4RxgLC"
   },
   "source": [
    "## LeNet-5 Architecture\n",
    "\n",
    "![Original Architecture of LeNet-5](https://miro.medium.com/max/4308/1*1TI1aGBZ4dybR6__DI9dzA.png)\n",
    "\n",
    "The original architecture of LeNet-5. In this homework, we use max instead of average pooling for subsampling, and use fully connected instead of gaussian final layer. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "tNfECWDixOHI"
   },
   "source": [
    "## PyTorch Implementation\n",
    "\n",
    "Please complete the definition of layers and forward pass in the following starter code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "EzJHjxSVMB1H"
   },
   "outputs": [],
   "source": [
    "class LeNet5(nn.Module):\n",
    "\n",
    "    # definition of each neural network layer\n",
    "    def __init__(self):\n",
    "        super(LeNet5, self).__init__()\n",
    "        self.C1 = nn.Conv2d(1, 6, kernel_size=(5, 5))\n",
    "        self.S2 = nn.MaxPool2d(kernel_size=(2, 2), stride=2)\n",
    "\n",
    "        ####### Complete the defition of C3 and S4 ##########\n",
    "        # C3 is a convolutional layer with 16 5x5 kernels\n",
    "        # S4 is a max pooling layer with 2x2 kernel and stride 2\n",
    "\n",
    "        self.C3 = nn.Conv2d(6,16,kernel_size=(5, 5))\n",
    "        self.S4 = nn.MaxPool2d(kernel_size=(2, 2), stride=2)\n",
    "        #####################################################\n",
    "\n",
    "        self.C5 = nn.Conv2d(16, 120, kernel_size=(5, 5))\n",
    "        self.F6 = nn.Linear(120, 84)\n",
    "\n",
    "        # output layer\n",
    "        self.OL = nn.Linear(84, 10)\n",
    "        \n",
    "        # record C5 & F6 activation for visualization\n",
    "        self.record = {\"C5\":None, \"F6\":None}\n",
    "\n",
    "    # definition of the forward pass\n",
    "    def forward(self, x):\n",
    "        \n",
    "        # input x are (batch, 1, 32, 32) grayscale images\n",
    "        # the first convolutional layer C1 with 6 kernels size 5×5 and a stride of 1\n",
    "        # output image size changes from (batch, 1, 32, 32) to (batch, 6, 28, 28)\n",
    "        # then pass the feature maps to the tanh activation function\n",
    "        x = torch.tanh(self.C1(x))\n",
    "\n",
    "        # pass the feature maps to a 2x2 max pooling layer S2 \n",
    "        # the output image dimension decreases halved -> (batch, 6, 14, 14)\n",
    "        x = self.S2(x)\n",
    "        \n",
    "        ####### Complete the foward pass C3 > S4 > C5 ###################################\n",
    "        # C3 is the second convolutional layer with 16 kernels size 5×5 and a stride of 1\n",
    "        # after C3, the output image size changes from (batch, 6, 14, 14) to (batch, 16, 10, 10)\n",
    "        # S4 is the second 2x2 pooling layer with strid 2\n",
    "        # after S4, the output image size changes from (batch, 16, 10, 10) to (batch, 16, 5, 5)\n",
    "        # C5 is the third convolutional layer with 120 kernels size 5×5 and a stride of 1\n",
    "        # after C5, the output image size changes from (batch, 16, 5, 5) to (batch, 120, 1, 1)\n",
    "        # C3, C5 is followed by tanh activations\n",
    "\n",
    "        x = torch.tanh(self.C3(x))\n",
    "        x = self.S4(x)\n",
    "        x = torch.tanh(self.C5(x))\n",
    "        ##################################################################################\n",
    "\n",
    "        # convert (batch, 120, 1, 1) feature maps to 1d features of size (batch, 120)\n",
    "        x = x.view(x.size(0), -1) \n",
    "        \n",
    "        # record the activation of C5 as a numpy array\n",
    "        # .detach() declares a tensor does not need gradients\n",
    "        # .numpy() convert a torch tensoer without gradient to numpy array\n",
    "        # independent of the forward pass, we have to make a copy of x by method .clone()\n",
    "        self.record[\"C5\"] = x.clone().detach().numpy()\n",
    "\n",
    "        # pass the activation to the fully connected layer F6 followed by a tanh activation\n",
    "        # output size changes from (batch, 120) to (batch, 86)\n",
    "        x = torch.tanh(self.F6(x))\n",
    "\n",
    "        # record the activat of F6 as numpy array\n",
    "        self.record[\"F6\"] = x.clone().detach().numpy()\n",
    "\n",
    "        # pass the activation to the final fully connected layer OL followed by a tanh activation\n",
    "        # output size changes from (batch, 86) to (batch, 10)\n",
    "        x = torch.tanh(self.OL(x))\n",
    "        \n",
    "        return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "TtyxZUX5f6_D"
   },
   "source": [
    "## Training with PyTorch\n",
    "\n",
    "Please read the following code carefully. Run the code to show that the learning curve goes down. (This code is written for pedagogical purposes, not efficiency.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "XvEkHlUmMB1J"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1332x756 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.887881278991699 seconds per 100 iterations\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1332x756 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ntrain = trainimages.shape[0];  # number of training examples\n",
    "nepoch = 10;                    # number of epochs through training set\n",
    "disp_freq = 100                 # display frequency\n",
    "batchsize = 32                  # minibatch size\n",
    "\n",
    "errs = []\n",
    "losses = []\n",
    "\n",
    "lenet5 = LeNet5()\n",
    "\n",
    "# use SGD optimizer, set learning rate parameter as 0.1\n",
    "optimizer = optim.SGD(lenet5.parameters(), lr=0.1)\n",
    "\n",
    "t_start = time.time()\n",
    "for iepoch in range(nepoch):\n",
    "    for t in range(int(ntrain / batchsize)):\n",
    "        batchindices = np.random.choice(ntrain, batchsize, replace=False)\n",
    "        trainlabels_iter = trainlabels[batchindices]\n",
    "        \n",
    "        # label 1 for the correct digit and -1 for the incorrect digits\n",
    "        y = torch.ones(10, batchsize) * (-1)\n",
    "        y[trainlabels_iter, torch.arange(batchsize, dtype=torch.int64)] = 1\n",
    "\n",
    "        # normalize input images\n",
    "        imgs = torch.zeros([batchsize, 1, 32, 32])\n",
    "        imgs[:, 0, 2: -2, 2: -2] = trainimages[batchindices].float() / 255.\n",
    "\n",
    "        # before the forward pass, clean the gradient buffers of all parameters\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        # forward pass\n",
    "        out = lenet5(imgs)\n",
    "        \n",
    "        # MSE loss\n",
    "        loss = torch.mean(0.5*(y - out.t())**2)\n",
    "\n",
    "        # backward pass\n",
    "        loss.backward()\n",
    "\n",
    "        # update parameters using SGD\n",
    "        optimizer.step()\n",
    "\n",
    "        # calculate error rate and loss for plot\n",
    "        pred = torch.argmax(out, dim=1)\n",
    "        err = torch.mean((pred != trainlabels_iter).float())\n",
    "        errs.append(err.detach().numpy())\n",
    "        losses.append(loss.detach().numpy())\n",
    "\n",
    "        \n",
    "        # plots\n",
    "        if (t + 1) % disp_freq == 0:\n",
    "            plt.gcf().clear()\n",
    "            visualize(len(errs) - 1, losses, errs, out[0,:].detach(), lenet5.record[\"F6\"][:, 0], \n",
    "                      lenet5.record[\"C5\"][:, 0], imgs[0, 0].detach(), lenet5.C1.weight.detach().squeeze())\n",
    "            print(str(time.time() - t_start) + \" seconds per \" + str(disp_freq) + \" iterations\")\n",
    "            t_start = time.time()\n",
    "            time.sleep(0.01)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xDy-y9XqivNA"
   },
   "source": [
    "### Use ReLU activation and change C5 to F5\n",
    "Now let's implement a varant of LeNet-5, use `ReLU` activation function instead of `tanh` activation function for each layer. And change the convolutional layer C5 in the original net work to a fully connected layer F5. You will have to use softmax as the network output. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YPlUN_KAivi7"
   },
   "outputs": [],
   "source": [
    "class LeNet5_Relu(nn.Module):\n",
    "\n",
    "    # definition of each neural network layer\n",
    "    def __init__(self):\n",
    "        super(LeNet5_Relu, self).__init__()\n",
    "        self.C1 = nn.Conv2d(1, 6, kernel_size=(5, 5))\n",
    "        self.S2 = nn.MaxPool2d(kernel_size=(2, 2), stride=2)\n",
    "        self.C3 = nn.Conv2d(6, 16, kernel_size=(5, 5))\n",
    "        self.S4 = nn.MaxPool2d(kernel_size=(2, 2), stride=2)\n",
    "        \n",
    "        ####### Complete the defition of F5 #################\n",
    "        \n",
    "        self.F5 = nn.linear(16,120)    \n",
    "        \n",
    "        #####################################################\n",
    "\n",
    "        self.F6 = nn.Linear(120, 84)\n",
    "\n",
    "        # output layer\n",
    "        self.OL = nn.Linear(84, 10)\n",
    "        \n",
    "        # record C5 & C6 activation for visualization\n",
    "        self.record = {\"F5\":None, \"F6\":None}\n",
    "\n",
    "    # definition of the forward pass\n",
    "    def forward(self, x):\n",
    "        \n",
    "        ####### Complete the foward pass C1 >> F5 ##########\n",
    "        \n",
    "        x = torch.tanh(self.C1(x))\n",
    "        x = self.S2(x)\n",
    "        x = torch.tanh(self.C3(x))\n",
    "        x = self.S4(x)\n",
    "        x = functional.relu(self.F5(x))\n",
    "        x = x.view(x.size(0), -1)      \n",
    "        \n",
    "        #####################################################\n",
    "\n",
    "        self.record[\"F5\"] = x.clone().detach().numpy()\n",
    "        x = torch.relu(self.F6(x))\n",
    "        self.record[\"F6\"] = x.clone().detach().numpy()\n",
    "        x = functional.softmax(self.OL(x), dim=1)\n",
    "        \n",
    "        return x"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "LeNet.ipynb",
   "provenance": []
  },
  "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}