CNNPYT.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dataset\n",
    "from __future__ import print_function\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "from torchvision import datasets, transforms\n",
    "from torch.autograd import Variable\n",
    "import numpy as np\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "import matplotlib.pyplot as plt\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "label_ls = ['Banana', 'Apple', 'Strawberry', 'Pineapple', 'Orange']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def PIL2Tensor(path,label,label_list):\n",
    "    \n",
    "    \"\"\"\n",
    "    This function takes in two String args\n",
    "    path : Absolute path to the image from dataframe\n",
    "    label : Corresponding Label to the image\n",
    "     Returns:\n",
    "     Image and Label as Variable \n",
    "    \"\"\"\n",
    "    image = Image.open(path)\n",
    "    image = transforms.ToTensor()(image).unsqueeze(0) # unsqueeze to add artificial first dimension\n",
    "    #image = Variable(image)\n",
    "    for i , x in enumerate(label_list):\n",
    "        if label == x:\n",
    "            break\n",
    "    return image , torch.tensor([i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = dataset.create_and_load_meta_csv_df(\"Data\",split=0.8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "class ImageLoad(Dataset):\n",
    "    \"\"\" Fruit dataset \"\"\"\n",
    "    \n",
    "    \n",
    "\n",
    "    # Initialize your data, download, etc.\n",
    "    def __init__(self,df):\n",
    "        \n",
    "        label = ['Banana', 'Apple', 'Strawberry', 'Pineapple', 'Orange'] #df.iloc[:,1].unique()\n",
    "        self.len = len(df.index)\n",
    "        self.x_data , self.y_data = PIL2Tensor(df.iloc[0,0],df.iloc[0,1],label)\n",
    "        \n",
    "        for co in range(self.len):\n",
    "            \n",
    "            temp_data , temp_label = PIL2Tensor(df.iloc[co,0],df.iloc[co,1],label)\n",
    "            self.x_data = torch.cat((self.x_data,temp_data))\n",
    "            self.y_data = torch.cat((self.y_data,temp_label))\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        return self.x_data[index], self.y_data[index]\n",
    "\n",
    "    def __len__(self):\n",
    "        return self.len"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Net(nn.Module):\n",
    "\n",
    "    def __init__(self):\n",
    "        super(Net, self).__init__()\n",
    "        self.conv1 = nn.Conv2d(3, 10, kernel_size=5)\n",
    "        self.conv2 = nn.Conv2d(10, 20, kernel_size=5)\n",
    "        self.mp = nn.MaxPool2d(2)\n",
    "        self.fc = nn.Linear(9680, 5)\n",
    "\n",
    "    def forward(self, x):\n",
    "        in_size = x.size(0)\n",
    "        x = F.relu(self.mp(self.conv1(x)))\n",
    "        x = F.relu(self.mp(self.conv2(x)))\n",
    "        x = x.view(in_size, -1)  # flatten the tensor\n",
    "        x = self.fc(x)\n",
    "        return F.log_softmax(x)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = ImageLoad(df[1])\n",
    "train_loader = DataLoader(dataset=dataset,\n",
    "                          batch_size=32,\n",
    "                          shuffle=True)\n",
    "dataset_train = ImageLoad(df[2])\n",
    "test_loader = DataLoader(dataset=dataset_train,\n",
    "                          batch_size=32,\n",
    "                          shuffle=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Net()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train(epoch):\n",
    "    model.train()\n",
    "    for batch_idx, (data, target) in enumerate(train_loader):\n",
    "        data, target = Variable(data), Variable(target)\n",
    "        optimizer.zero_grad()\n",
    "        output = model(data)\n",
    "        loss = F.nll_loss(output, target)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        if batch_idx % 10 == 0:\n",
    "            print('Train Epoch: {} [{}/{} ({:.0f}%)]\\tLoss: {:.6f}'.format(\n",
    "                epoch, batch_idx * len(data), len(train_loader.dataset),\n",
    "                100. * batch_idx / len(train_loader), loss.data[0]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def test():\n",
    "    model.eval()\n",
    "    test_loss = 0\n",
    "    correct = 0\n",
    "    for data, target in test_loader:\n",
    "        data, target = Variable(data, volatile=True), Variable(target)\n",
    "        output = model(data)\n",
    "        # sum up batch loss\n",
    "        test_loss += F.nll_loss(output, target, size_average=False).data[0]\n",
    "        # get the index of the max log-probability\n",
    "        pred = output.data.max(1, keepdim=True)[1]\n",
    "        correct += pred.eq(target.data.view_as(pred)).cpu().sum()\n",
    "\n",
    "    test_loss /= len(test_loader.dataset)\n",
    "    print('\\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\\n'.format(\n",
    "        test_loss, correct, len(test_loader.dataset),\n",
    "        100. * correct / len(test_loader.dataset)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sameer/anaconda3/envs/HC#4705_stage1/lib/python3.6/site-packages/ipykernel_launcher.py:16: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.\n",
      "  app.launch_new_instance()\n",
      "/home/sameer/anaconda3/envs/HC#4705_stage1/lib/python3.6/site-packages/ipykernel_launcher.py:13: UserWarning: invalid index of a 0-dim tensor. This will be an error in PyTorch 0.5. Use tensor.item() to convert a 0-dim tensor to a Python number\n",
      "  del sys.path[0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Epoch: 1 [0/1954 (0%)]\tLoss: 1.598716\n",
      "Train Epoch: 1 [320/1954 (16%)]\tLoss: 1.510603\n",
      "Train Epoch: 1 [640/1954 (32%)]\tLoss: 0.925762\n",
      "Train Epoch: 1 [960/1954 (48%)]\tLoss: 0.615894\n",
      "Train Epoch: 1 [1280/1954 (65%)]\tLoss: 0.151429\n",
      "Train Epoch: 1 [1600/1954 (81%)]\tLoss: 0.694701\n",
      "Train Epoch: 1 [1920/1954 (97%)]\tLoss: 0.077839\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sameer/anaconda3/envs/HC#4705_stage1/lib/python3.6/site-packages/ipykernel_launcher.py:6: UserWarning: volatile was removed and now has no effect. Use `with torch.no_grad():` instead.\n",
      "  \n",
      "/home/sameer/anaconda3/envs/HC#4705_stage1/lib/python3.6/site-packages/torch/nn/functional.py:52: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.\n",
      "  warnings.warn(warning.format(ret))\n",
      "/home/sameer/anaconda3/envs/HC#4705_stage1/lib/python3.6/site-packages/ipykernel_launcher.py:9: UserWarning: invalid index of a 0-dim tensor. This will be an error in PyTorch 0.5. Use tensor.item() to convert a 0-dim tensor to a Python number\n",
      "  if __name__ == '__main__':\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Test set: Average loss: 1.0865, Accuracy: 358/489 (73%)\n",
      "\n",
      "Train Epoch: 2 [0/1954 (0%)]\tLoss: 0.907558\n",
      "Train Epoch: 2 [320/1954 (16%)]\tLoss: 0.040669\n",
      "Train Epoch: 2 [640/1954 (32%)]\tLoss: 0.019708\n",
      "Train Epoch: 2 [960/1954 (48%)]\tLoss: 0.032243\n",
      "Train Epoch: 2 [1280/1954 (65%)]\tLoss: 0.025043\n",
      "Train Epoch: 2 [1600/1954 (81%)]\tLoss: 0.014207\n",
      "Train Epoch: 2 [1920/1954 (97%)]\tLoss: 0.008167\n",
      "\n",
      "Test set: Average loss: 0.0206, Accuracy: 486/489 (99%)\n",
      "\n",
      "Train Epoch: 3 [0/1954 (0%)]\tLoss: 0.035760\n",
      "Train Epoch: 3 [320/1954 (16%)]\tLoss: 0.112360\n",
      "Train Epoch: 3 [640/1954 (32%)]\tLoss: 0.005757\n",
      "Train Epoch: 3 [960/1954 (48%)]\tLoss: 0.008149\n",
      "Train Epoch: 3 [1280/1954 (65%)]\tLoss: 0.009434\n",
      "Train Epoch: 3 [1600/1954 (81%)]\tLoss: 0.023403\n",
      "Train Epoch: 3 [1920/1954 (97%)]\tLoss: 0.005780\n",
      "\n",
      "Test set: Average loss: 0.0082, Accuracy: 489/489 (100%)\n",
      "\n",
      "Train Epoch: 4 [0/1954 (0%)]\tLoss: 0.007484\n",
      "Train Epoch: 4 [320/1954 (16%)]\tLoss: 0.002090\n",
      "Train Epoch: 4 [640/1954 (32%)]\tLoss: 0.008415\n",
      "Train Epoch: 4 [960/1954 (48%)]\tLoss: 0.004911\n",
      "Train Epoch: 4 [1280/1954 (65%)]\tLoss: 0.006314\n",
      "Train Epoch: 4 [1600/1954 (81%)]\tLoss: 0.001159\n",
      "Train Epoch: 4 [1920/1954 (97%)]\tLoss: 0.002852\n",
      "\n",
      "Test set: Average loss: 0.0038, Accuracy: 489/489 (100%)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for epoch in range(1, 5):\n",
    "    train(epoch)\n",
    "    test()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict(df,i):\n",
    "    \"\"\"\n",
    "    Gives the prediction for ith sample from th Dataframe\n",
    "    \"\"\"\n",
    "    image = Image.open(df.iloc[i,0])\n",
    "    plt.imshow(image)\n",
    "    lab_name = df.iloc[i,1]\n",
    "    plt.title(lab_name)\n",
    "    image = transforms.ToTensor()(image).unsqueeze(0) # unsqueeze to add artificial first dimension\n",
    "    pred_lab = model(Variable(image))\n",
    "    for x , y in enumerate(label_ls):\n",
    "        if y ==lab_name:\n",
    "            break\n",
    "    print(torch.argmax(pred_lab[0,:]),x)\n",
    "    print(label_ls)\n",
    "    return None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor(4) 4\n",
      "['Banana', 'Apple', 'Strawberry', 'Pineapple', 'Orange']\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sameer/anaconda3/envs/HC#4705_stage1/lib/python3.6/site-packages/ipykernel_launcher.py:16: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.\n",
      "  app.launch_new_instance()\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "predict(df[0],90)"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}