{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Pytorch Handwritten Digit Prediction","execution_count":null},{"metadata":{},"cell_type":"markdown","source":"# 1. Import Libraries","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nfrom time import time\n\nimport torch\nfrom torch import optim\nimport torch.nn as nn\nimport torch.nn.functional as F\n\nfrom sklearn.preprocessing import MinMaxScaler","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 2. Load Data","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')\ntest = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')\n\ntrain_set = np.array(train.drop(['label'], axis=1))\ntrain_label = np.array(train['label'])","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 3. Preprocess Data","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"trainset = []\nbatch_size = 64\nvalid_size = 0.2\n\nfor i in range(len(train_label)):    \n    trainset.append((torch.tensor(train_set[i]), torch.tensor(train_label[i])))\n\nvalid_num = int(len(trainset)*0.2)\ntrainloader = torch.utils.data.DataLoader(trainset[valid_num:], batch_size=64, shuffle=True)\nvalidloader = torch.utils.data.DataLoader(trainset[:valid_num], batch_size=64, shuffle=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"dataiter = iter(trainloader)\nimages, labels = dataiter.next()\nprint(images.shape)\nprint(labels.shape)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"img = images[3].reshape(1,28,28)\nplt.imshow(img.numpy().squeeze(), cmap='gray_r')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"figure = plt.figure(figsize=(10,10))\nnum_of_images = 60\nfor i in range(0, num_of_images):\n    plt.subplot(6,10,i+1)\n    img = images[i].reshape(1,28,28)\n    plt.imshow(img.numpy().squeeze(), cmap='gray_r')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 3. Build Model","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"class Net(nn.Module):\n    def __init__(self):\n        super(Net, self).__init__()\n        self.fc1 = nn.Linear(784, 512)\n        self.fc2 = nn.Linear(512, 256)\n        self.fc3 = nn.Linear(256, 10)\n        self.dropout = nn.Dropout(0.2)\n        \n    def forward(self, input):\n        x = F.relu(self.fc1(input))\n        x = self.dropout(x)\n        x = F.relu(self.fc2(x))\n        x = self.dropout(x)\n        x = F.log_softmax(self.fc3(x), dim=1)\n        return x","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 4. Train Model","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"model = Net()\n\ncriterion = nn.NLLLoss()\noptimizer = optim.SGD(model.parameters(), lr=0.003, momentum=0.9)\n\nn_epochs = 30\ntime0 = time()\nscaler = MinMaxScaler((0,1))\n\n\nfor epoch in range(n_epochs):\n    train_loss = 0.0\n    valid_loss = 0.0\n    \n    for images, labels in trainloader:\n        optimizer.zero_grad() \n        images_scaling = scaler.fit_transform(images.view(-1,len(images)))\n        output = model(torch.tensor(images_scaling).view(len(images),-1).float())\n        loss = criterion(output, labels)\n        loss.backward()\n        optimizer.step()\n        train_loss += loss.item()*images.size(0)\n        \n    for images, labels in validloader:\n        images_scaling = scaler.fit_transform(images.view(-1, len(images)))\n        output = model(torch.tensor(images_scaling).view(len(images), -1).float())\n        loss = criterion(output, labels)\n        valid_loss += loss.item()*images.size(0)\n    \n    train_loss = train_loss/len(trainset[valid_num:])\n    valid_loss = valid_loss/len(trainset[:valid_num])\n    print('Epoch: {} \\tTraining loss: {:.6f} \\tValidation loss: {:.6f}'.format(\n        epoch+1,\n        train_loss,\n        valid_loss\n    ))\n    \n#     if valid_loss <= valid_loss_min\nprint(\"\\nTraining Time (in minutes) =\",(time()-time0)/60)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 5. Test and Evaluation","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"images, labels = next(iter(validloader))\n\nimg = images[1]\nimages_scaling = scaler.fit_transform(img.reshape(-1,1)).reshape(1,-1)\n\nwith torch.no_grad():\n    logps = model(torch.from_numpy(images_scaling).float())\n\nps = torch.exp(logps)\nprobab = list(ps.numpy()[0])\n\nprint(\"Predicted Digit =\", probab.index(max(probab)))\nprint(\"Label Digit =\", labels[1])\n\nimg = img.reshape(1,28,28)\nplt.imshow(img.numpy().squeeze(), cmap='gray_r')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"correct_count, all_count = 0, 0\nfor images,labels in validloader:\n    for i in range(len(labels)):\n        img = images[i]\n        images_scaling = scaler.fit_transform(img.reshape(-1,1)).reshape(1,-1) \n        with torch.no_grad():\n            logps = model(torch.from_numpy(images_scaling).float())\n\n        ps = torch.exp(logps)\n        probab = list(ps.numpy()[0])\n        pred_label = probab.index(max(probab))\n        true_label = labels.numpy()[i]\n        if(true_label == pred_label):\n          correct_count += 1\n        all_count += 1\n\nprint(\"Number Of Images Tested =\", all_count)\nprint(\"\\nModel Accuracy =\", (correct_count/all_count))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 6. Predict Test Data","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"test_set = np.array(test)\ntestloader = torch.utils.data.DataLoader(test_set, batch_size=64, shuffle=False)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pred_test = []\npred_id = []\nimageid = 1\nfor images in testloader:\n    for i in range(len(images)):\n        img = images[i]\n        images_scaling = scaler.fit_transform(img.reshape(-1,1)).reshape(1,-1) \n        with torch.no_grad():\n            logps = model(torch.from_numpy(images_scaling).float())\n\n        ps = torch.exp(logps)\n        probab = list(ps.numpy()[0])\n        pred_label = probab.index(max(probab))\n        pred_test.append(pred_label)\n        pred_id.append(imageid)\n        imageid += 1","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"my_prediction = pd.DataFrame({'ImageId' : pred_id, 'Label' : pred_test})\nmy_prediction.to_csv('my_prediction.csv', index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}