{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":16017,"databundleVersionId":708189,"sourceType":"competition"}],"dockerImageVersionId":30636,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-01-12T16:07:53.633433Z","iopub.execute_input":"2024-01-12T16:07:53.633792Z","iopub.status.idle":"2024-01-12T16:07:53.642499Z","shell.execute_reply.started":"2024-01-12T16:07:53.633758Z","shell.execute_reply":"2024-01-12T16:07:53.641448Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Ex 3: Train a model to classify Kanna digit Recognition","metadata":{}},{"cell_type":"markdown","source":"### 1. Importing library","metadata":{}},{"cell_type":"code","source":"# import libraries\nimport torch\nfrom torch.utils.data import DataLoader, Dataset\nimport torchvision\nimport torchvision.transforms as transforms\nfrom torch.utils.data.sampler import SubsetRandomSampler\nimport matplotlib.pyplot as plt\nimport numpy as np\nimport pandas as pd","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:53.643872Z","iopub.execute_input":"2024-01-12T16:07:53.644286Z","iopub.status.idle":"2024-01-12T16:07:53.651789Z","shell.execute_reply.started":"2024-01-12T16:07:53.644251Z","shell.execute_reply":"2024-01-12T16:07:53.650824Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 2. Define Custom dataset class\n\nNguyên tắc: 1 custom dataset class bao gồm cũng có 3 methods:\n\n1. __getitem__ : trả về data (và/hoặc label) khi biết số index\n2. __len__ : trả về số mẫu trong dataset\n3. __init__: thực hiện các thao tác trong khi khởi tạo 1 object dataset","metadata":{}},{"cell_type":"code","source":"class TrainMNIST(Dataset):\n    \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        # load image as ndarray type (Height * Width * Channels)\n        # be carefull for converting dtype to np.uint8 [Unsigned integer (0 to 255)]\n        # in this example, we use ToTensor(), so we define the numpy array like (H, W, C)\n        image = self.data.iloc[index, 1:].values.astype(np.uint8).reshape((28, 28, 1))\n        label = self.data.iloc[index, 0]\n        \n        if self.transform is not None:\n            image = self.transform(image)\n            \n        return image, label","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:53.653871Z","iopub.execute_input":"2024-01-12T16:07:53.654226Z","iopub.status.idle":"2024-01-12T16:07:53.664546Z","shell.execute_reply.started":"2024-01-12T16:07:53.654194Z","shell.execute_reply":"2024-01-12T16:07:53.663545Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class TestMNIST(Dataset):\n    \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        # load image as ndarray type (Height * Width * Channels)\n        # be carefull for converting dtype to np.uint8 [Unsigned integer (0 to 255)]\n        # in this example, we use ToTensor(), so we define the numpy array like (H, W, C)\n        image = self.data.iloc[index, 1:].values.astype(np.uint8).reshape((28, 28, 1))\n    \n        \n        if self.transform is not None:\n            image = self.transform(image)\n            \n        return image","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:53.665929Z","iopub.execute_input":"2024-01-12T16:07:53.666288Z","iopub.status.idle":"2024-01-12T16:07:53.676915Z","shell.execute_reply.started":"2024-01-12T16:07:53.666255Z","shell.execute_reply":"2024-01-12T16:07:53.675889Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"---\n### 3. Load and Visualize the [Data](http://pytorch.org/docs/stable/torchvision/datasets.html)\n\nThử dùng 2 class dataset cho 2 tập test và train ở trên để visualize 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":{"execution":{"iopub.status.busy":"2024-01-12T16:07:53.679192Z","iopub.execute_input":"2024-01-12T16:07:53.679516Z","iopub.status.idle":"2024-01-12T16:07:57.259688Z","shell.execute_reply.started":"2024-01-12T16:07:53.679485Z","shell.execute_reply":"2024-01-12T16:07:57.258698Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# number of subprocesses to use for data loading\nnum_workers = 0\n# how many samples per batch to load\nbatch_size = 32\n\n# convert data to torch.FloatTensor\n#transform = transforms.ToTensor()\n\n\n# prepare data loaders\ntrain_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size,\n     num_workers=num_workers,)\ntest_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, \n    num_workers=num_workers)","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:57.260825Z","iopub.execute_input":"2024-01-12T16:07:57.261113Z","iopub.status.idle":"2024-01-12T16:07:57.267188Z","shell.execute_reply.started":"2024-01-12T16:07:57.261087Z","shell.execute_reply":"2024-01-12T16:07:57.266234Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n%matplotlib inline\n    \n# obtain one batch of training images\ndataiter = iter(train_loader)\nimages, labels = next(dataiter)\nimages = images.numpy()\n\n# plot the images in the batch, along with the corresponding labels\nfig = plt.figure(figsize=(25, 4))\nfor idx in np.arange(20):\n    ax = fig.add_subplot(2, int(20/2), idx+1, xticks=[], yticks=[])\n    ax.imshow(np.squeeze(images[idx]), cmap='gray')\n    # print out the correct label for each image\n    # .item() gets the value contained in a Tensor\n    ax.set_title(str(labels[idx].item()))","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:57.268493Z","iopub.execute_input":"2024-01-12T16:07:57.268808Z","iopub.status.idle":"2024-01-12T16:07:58.442709Z","shell.execute_reply.started":"2024-01-12T16:07:57.268776Z","shell.execute_reply":"2024-01-12T16:07:58.441690Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"---\n### 4. Thiết kế 1 mạng NN đơn giản [Architecture](http://pytorch.org/docs/stable/nn.html)\n\nThiết kế 1 mạng neural network đơn giản cho việc phân loại","metadata":{}},{"cell_type":"code","source":"import torch.nn as nn\nimport torch.nn.functional as F\n\n# define the NN architecture\nclass Net(nn.Module):\n    def __init__(self):\n        super(Net, self).__init__()\n        # number of hidden nodes in each layer (512)\n        hidden_1 = 512\n        hidden_2 = 512\n        # linear layer (784 -> hidden_1)\n        \n        self.fc1 = nn.Linear(28*28, hidden_1)\n        # linear layer (n_hidden -> hidden_2)\n        self.fc2 = nn.Linear(hidden_1, hidden_2)\n        # linear layer (n_hidden -> 10)\n        self.fc3 = nn.Linear(hidden_2, 10)\n\n\n    def forward(self, x):\n        # flatten image input\n        x = x.view(-1, 28 * 28)\n        x = F.relu(self.fc1(x))\n        x = F.relu(self.fc2(x))\n        x = self.fc3(x)\n        return x\n\n# initialize the NN\ndevice = torch.device(\"cuda\")\n\nmodel = Net().to(device)\nprint(model)","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:58.443769Z","iopub.execute_input":"2024-01-12T16:07:58.444027Z","iopub.status.idle":"2024-01-12T16:07:58.460160Z","shell.execute_reply.started":"2024-01-12T16:07:58.444004Z","shell.execute_reply":"2024-01-12T16:07:58.459204Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model(torch.zeros((20,28,28,1)).to(device))","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:58.461166Z","iopub.execute_input":"2024-01-12T16:07:58.461435Z","iopub.status.idle":"2024-01-12T16:07:58.474482Z","shell.execute_reply.started":"2024-01-12T16:07:58.461412Z","shell.execute_reply":"2024-01-12T16:07:58.473659Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"###  5. Chọn [Loss Function](http://pytorch.org/docs/stable/nn.html#loss-functions) và [Optimizer](http://pytorch.org/docs/stable/optim.html)","metadata":{}},{"cell_type":"code","source":"# specify loss function (categorical cross-entropy)\ncriterion = nn.CrossEntropyLoss()\n\n# specify optimizer (stochastic gradient descent) and learning rate = 0.01\noptimizer = torch.optim.Adam(model.parameters(), lr=0.001)","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:07:58.475446Z","iopub.execute_input":"2024-01-12T16:07:58.475695Z","iopub.status.idle":"2024-01-12T16:07:58.480580Z","shell.execute_reply.started":"2024-01-12T16:07:58.475671Z","shell.execute_reply":"2024-01-12T16:07:58.479720Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"---\n### 6. Train the Network","metadata":{}},{"cell_type":"code","source":"from tqdm import tqdm\n# number of epochs to train the model\nn_epochs = 20\n\n# initialize tracker for minimum validation loss\ntrain_loss_min = np.Inf # set initial \"min\" to infinity\n\nfor epoch in range(n_epochs):\n    # monitor training loss\n    train_loss = 0.0\n    \n    ###################\n    # train the model #\n    ###################\n    model.train() # prep model for training\n    for data, target in tqdm(train_loader):\n        # clear the gradients of all optimized variables\n        optimizer.zero_grad()\n        # move to gpu\n        data = data.to(device)\n        target = target.to(device)\n        # forward pass: compute predicted outputs by passing inputs to the model\n        output = model(data)\n        # calculate the loss\n        loss = criterion(output, target)\n        # backward pass: compute gradient of the loss with respect to model parameters\n        loss.backward()\n        # perform a single optimization step (parameter update)\n        optimizer.step()\n        # update running training loss\n        train_loss += loss.item()*data.size(0)\n        \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":{"execution":{"iopub.status.busy":"2024-01-12T16:07:58.481537Z","iopub.execute_input":"2024-01-12T16:07:58.481781Z","iopub.status.idle":"2024-01-12T16:13:04.292969Z","shell.execute_reply.started":"2024-01-12T16:07:58.481758Z","shell.execute_reply":"2024-01-12T16:13:04.292030Z"},"trusted":true},"execution_count":null,"outputs":[]},{"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":{"execution":{"iopub.status.busy":"2024-01-12T16:13:04.294495Z","iopub.execute_input":"2024-01-12T16:13:04.294865Z","iopub.status.idle":"2024-01-12T16:13:04.305289Z","shell.execute_reply.started":"2024-01-12T16:13:04.294828Z","shell.execute_reply":"2024-01-12T16:13:04.304322Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"---\n## Test the Trained Network\n","metadata":{}},{"cell_type":"markdown","source":"### 7. Predict digit dựa vào input >> tạo file submission trong kaggle","metadata":{}},{"cell_type":"code","source":"test_loader = torch.utils.data.DataLoader(test_data, batch_size=1, \n    num_workers=num_workers)\n","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:13:04.306577Z","iopub.execute_input":"2024-01-12T16:13:04.306955Z","iopub.status.idle":"2024-01-12T16:13:04.311830Z","shell.execute_reply.started":"2024-01-12T16:13:04.306931Z","shell.execute_reply":"2024-01-12T16:13:04.310938Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"\ndef predict(model, dataloader):\n    prediction_list = []\n    for i, batch in enumerate(dataloader):\n        data = batch\n        data = data.to(device)\n        outputs = model(data)\n        _, predicted = torch.max(outputs.data, 1) \n        prediction_list.append(predicted.cpu())\n    return prediction_list","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:13:04.312903Z","iopub.execute_input":"2024-01-12T16:13:04.313151Z","iopub.status.idle":"2024-01-12T16:13:04.324533Z","shell.execute_reply.started":"2024-01-12T16:13:04.313128Z","shell.execute_reply":"2024-01-12T16:13:04.323358Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"predictions = predict(model,test_loader)\nvalues = [t.item() for t in predictions]\npredictions = np.array(values)","metadata":{"execution":{"iopub.status.busy":"2024-01-12T16:13:04.325723Z","iopub.execute_input":"2024-01-12T16:13:04.326016Z","iopub.status.idle":"2024-01-12T16:13:07.326953Z","shell.execute_reply.started":"2024-01-12T16:13:04.325993Z","shell.execute_reply":"2024-01-12T16:13:07.325998Z"},"trusted":true},"execution_count":null,"outputs":[]},{"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":{"execution":{"iopub.status.busy":"2024-01-12T16:13:07.328718Z","iopub.execute_input":"2024-01-12T16:13:07.329001Z","iopub.status.idle":"2024-01-12T16:13:07.654210Z","shell.execute_reply.started":"2024-01-12T16:13:07.328975Z","shell.execute_reply":"2024-01-12T16:13:07.653241Z"},"trusted":true},"execution_count":null,"outputs":[]}]}