{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":16017,"databundleVersionId":708189,"sourceType":"competition"}],"dockerImageVersionId":30786,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# 1. Importing library","metadata":{}},{"cell_type":"code","source":"# Import library\n\nimport torch\nfrom torch.utils.data import DataLoader, Dataset\nimport torch\nimport torchvision.transforms as transforms\nfrom torch.utils.data.sampler import SubsetRandomSampler\nimport matplotlib.pyplot as plt\nimport pandas as pd\nimport numpy as np","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:21.123495Z","iopub.execute_input":"2024-12-02T05:40:21.123959Z","iopub.status.idle":"2024-12-02T05:40:25.864612Z","shell.execute_reply.started":"2024-12-02T05:40:21.123923Z","shell.execute_reply":"2024-12-02T05:40:25.863483Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# 2. Define Custom dataset class","metadata":{}},{"cell_type":"code","source":"class TrainMNIST(Dataset):\n    def __init__(self, file_path, transform = None):\n        self.data = pd.read_csv(file_path)\n        self.transform = transform\n        \n    def __len__(self):\n        return len(self.data)\n\n    def __getitem__(self, index):\n        image = self.data.iloc[index, 1:].values.astype(np.uint8).reshape((28,28,1))\n        label = self.data.iloc[index, 0]\n        if self.transform is not None:\n            image = self.transform(image)\n        return image, label    \n\nclass TestMNIST(Dataset):\n    def __init__(self, file_path, transform = None):\n        self.data = pd.read_csv(file_path)\n        self.transform = transform\n        \n    def __len__(self):\n        return len(self.data)\n\n    def __getitem__(self, index):\n        image = self.data.iloc[index, 1:].values.astype(np.uint8).reshape((28,28,1))\n        if self.transform is not None:\n            image = self.transform(image)\n        return image","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:25.866421Z","iopub.execute_input":"2024-12-02T05:40:25.867477Z","iopub.status.idle":"2024-12-02T05:40:25.875708Z","shell.execute_reply.started":"2024-12-02T05:40:25.867441Z","shell.execute_reply":"2024-12-02T05:40:25.874480Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# 3. Load and Visualize the Data","metadata":{}},{"cell_type":"code","source":"train_data = TrainMNIST('/kaggle/input/Kannada-MNIST/train.csv', transform=transforms.ToTensor())\ntest_data = TestMNIST('/kaggle/input/Kannada-MNIST/test.csv', transform=transforms.ToTensor())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:25.877173Z","iopub.execute_input":"2024-12-02T05:40:25.877629Z","iopub.status.idle":"2024-12-02T05:40:30.650680Z","shell.execute_reply.started":"2024-12-02T05:40:25.877596Z","shell.execute_reply":"2024-12-02T05:40:30.649629Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_data.__len__()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:30.653224Z","iopub.execute_input":"2024-12-02T05:40:30.653589Z","iopub.status.idle":"2024-12-02T05:40:30.661434Z","shell.execute_reply.started":"2024-12-02T05:40:30.653555Z","shell.execute_reply":"2024-12-02T05:40:30.660227Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"test_data.__len__()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:30.662549Z","iopub.execute_input":"2024-12-02T05:40:30.662870Z","iopub.status.idle":"2024-12-02T05:40:30.673773Z","shell.execute_reply.started":"2024-12-02T05:40:30.662840Z","shell.execute_reply":"2024-12-02T05:40:30.672800Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"num_workers = 0\n\nbatch_size = 20\n\n# transform = transforms.ToTensor()\ntrain_loader = torch.utils.data.DataLoader(train_data, batch_size = batch_size, num_workers = num_workers)\ntest_loader = torch.utils.data.DataLoader(test_data, batch_size = batch_size, num_workers = num_workers)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:30.674991Z","iopub.execute_input":"2024-12-02T05:40:30.675316Z","iopub.status.idle":"2024-12-02T05:40:30.685002Z","shell.execute_reply.started":"2024-12-02T05:40:30.675265Z","shell.execute_reply":"2024-12-02T05:40:30.683905Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\ndataiter = iter(train_loader)\n\nimages, labels = next(dataiter)\nimages = images.numpy()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:30.686579Z","iopub.execute_input":"2024-12-02T05:40:30.687012Z","iopub.status.idle":"2024-12-02T05:40:30.788180Z","shell.execute_reply.started":"2024-12-02T05:40:30.686967Z","shell.execute_reply":"2024-12-02T05:40:30.786936Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"fig = plt.figure(figsize = (25, 4))\nfor idx in np.arange(20):\n    ax = fig.add_subplot(2, round(20/2), idx+1, xticks = [], yticks = [])\n    ax.imshow(np.squeeze(images[idx]), cmap = 'gray')\n    ax.set_title(str(labels[idx].item()))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:30.789706Z","iopub.execute_input":"2024-12-02T05:40:30.790129Z","iopub.status.idle":"2024-12-02T05:40:31.846444Z","shell.execute_reply.started":"2024-12-02T05:40:30.790085Z","shell.execute_reply":"2024-12-02T05:40:31.845286Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# 4. Thiết kế mạng neutral network","metadata":{}},{"cell_type":"code","source":"import torch.nn as nn\nimport torch.nn.functional as F\n\nclass Net(nn.Module):\n    def __init__(self):\n        # number of hidden nodes in each layer\n        super(Net, self).__init__()\n        hidden_1 = 512\n        hidden_2 = 512\n        # linear layer\n        self.fc1 = nn.Linear(28*28, hidden_1)\n        self.fc2 = nn.Linear(hidden_1, hidden_2)\n        self.fc3 = nn.Linear(hidden_2, 10)\n    def forward(self, x):\n        x = x.view(-1, 28*28)\n        # add hidden layer, with relu activation function\n        x = F.relu(self.fc1(x))\n        x = F.relu(self.fc2(x))\n        x = self.fc3(x)\n        return x","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:31.847792Z","iopub.execute_input":"2024-12-02T05:40:31.848087Z","iopub.status.idle":"2024-12-02T05:40:31.855353Z","shell.execute_reply.started":"2024-12-02T05:40:31.848058Z","shell.execute_reply":"2024-12-02T05:40:31.854198Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# khai báo mô hình\n# device=torch.device(\"cuda\")\nmodel = Net()\nprint(model)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:31.858426Z","iopub.execute_input":"2024-12-02T05:40:31.858877Z","iopub.status.idle":"2024-12-02T05:40:31.877078Z","shell.execute_reply.started":"2024-12-02T05:40:31.858844Z","shell.execute_reply":"2024-12-02T05:40:31.875829Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# model(torch.zeros((20,28,28,1)).to(device))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:31.878525Z","iopub.execute_input":"2024-12-02T05:40:31.878950Z","iopub.status.idle":"2024-12-02T05:40:31.884001Z","shell.execute_reply.started":"2024-12-02T05:40:31.878903Z","shell.execute_reply":"2024-12-02T05:40:31.882713Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# 5. Chọn Loss Function và Optimizer","metadata":{}},{"cell_type":"code","source":"# Khai báo loss function\n\ncriterion = nn.CrossEntropyLoss()\n\noptimizer = torch.optim.SGD(model.parameters(), lr = 0.01)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:31.885930Z","iopub.execute_input":"2024-12-02T05:40:31.886367Z","iopub.status.idle":"2024-12-02T05:40:31.895708Z","shell.execute_reply.started":"2024-12-02T05:40:31.886322Z","shell.execute_reply":"2024-12-02T05:40:31.894585Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# 6. Train the Network","metadata":{}},{"cell_type":"code","source":"# from tqdm import tqdm\n\nn_epochs = 20\n\ntrain_loss_min = np.Inf\n\nfor epoch in range(n_epochs):\n    train_loss = 0.0\n    # train the model\n    model.train()\n    for data, target in train_loader:\n        optimizer.zero_grad()\n\n        # data = data.to(device)\n        # target = target.to(device)\n        # forward pass\n        output = model(data)\n        # calculate the loss \n        loss = criterion(output, target)\n        # barkward pass\n        loss.backward()\n        # parameter update\n        optimizer.step()\n        # update running training loss\n        train_loss += loss.item()*data.size(0)\n\n    train_loss = train_loss/len(train_loader.dataset)\n    \n    print('Epoch: {} \\tTraining Loss: {:.6f}'.format(\n        epoch+1, \n        train_loss,\n        ))\n    \n    # save model if validation loss has decreased\n    if train_loss <= train_loss_min:\n        print('Training loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(\n        train_loss_min,\n        train_loss))\n        torch.save(model.state_dict(), 'model.pt')\n        train_loss_min = train_loss","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:40:31.897078Z","iopub.execute_input":"2024-12-02T05:40:31.897506Z","iopub.status.idle":"2024-12-02T05:46:26.453624Z","shell.execute_reply.started":"2024-12-02T05:40:31.897453Z","shell.execute_reply":"2024-12-02T05:46:26.452470Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Load the Model with the Lowest Validation loss","metadata":{}},{"cell_type":"code","source":"model.load_state_dict(torch.load('model.pt'))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:46:26.455059Z","iopub.execute_input":"2024-12-02T05:46:26.455706Z","iopub.status.idle":"2024-12-02T05:46:26.466873Z","shell.execute_reply.started":"2024-12-02T05:46:26.455658Z","shell.execute_reply":"2024-12-02T05:46:26.465790Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Test the Trained Network","metadata":{}},{"cell_type":"code","source":"test_loader = torch.utils.data.DataLoader(test_data, batch_size=1, \n    num_workers=num_workers)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:46:26.468298Z","iopub.execute_input":"2024-12-02T05:46:26.468849Z","iopub.status.idle":"2024-12-02T05:46:26.474294Z","shell.execute_reply.started":"2024-12-02T05:46:26.468802Z","shell.execute_reply":"2024-12-02T05:46:26.473271Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# 7. Dự đoán ","metadata":{}},{"cell_type":"code","source":"def predict(model, dataloader):\n    prediction_list = []\n    for i, batch in enumerate(dataloader):\n        outputs = model(batch)\n        _, predicted = torch.max(outputs.data, 1) \n        prediction_list.append(predicted.cpu())\n    return prediction_list","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:46:26.475504Z","iopub.execute_input":"2024-12-02T05:46:26.475909Z","iopub.status.idle":"2024-12-02T05:46:26.487351Z","shell.execute_reply.started":"2024-12-02T05:46:26.475877Z","shell.execute_reply":"2024-12-02T05:46:26.486323Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"predictions = predict(model,test_loader)\npredictions = np.array(predictions)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:46:26.488791Z","iopub.execute_input":"2024-12-02T05:46:26.489117Z","iopub.status.idle":"2024-12-02T05:46:29.324378Z","shell.execute_reply.started":"2024-12-02T05:46:26.489086Z","shell.execute_reply":"2024-12-02T05:46:29.323284Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"predictions = predictions.reshape(-1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:46:29.326998Z","iopub.execute_input":"2024-12-02T05:46:29.327345Z","iopub.status.idle":"2024-12-02T05:46:29.332290Z","shell.execute_reply.started":"2024-12-02T05:46:29.327313Z","shell.execute_reply":"2024-12-02T05:46:29.331131Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"test_data = pd.read_csv(\"/kaggle/input/Kannada-MNIST/test.csv\")\n\n\nsubmission = pd.DataFrame(data={\n    \"id\": test_data.index,\n    \"label\": predictions\n})\nsubmission.to_csv(\"submission.csv\", index=None)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:46:29.333653Z","iopub.execute_input":"2024-12-02T05:46:29.333974Z","iopub.status.idle":"2024-12-02T05:46:29.618592Z","shell.execute_reply.started":"2024-12-02T05:46:29.333944Z","shell.execute_reply":"2024-12-02T05:46:29.617685Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"pd.read_csv('/kaggle/working/submission.csv')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T05:47:27.702811Z","iopub.execute_input":"2024-12-02T05:47:27.703217Z","iopub.status.idle":"2024-12-02T05:47:27.728501Z","shell.execute_reply.started":"2024-12-02T05:47:27.703182Z","shell.execute_reply":"2024-12-02T05:47:27.727089Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}