{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":3004,"databundleVersionId":861823,"sourceType":"competition"}],"dockerImageVersionId":30746,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"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":"2024-07-28T12:57:50.592251Z","iopub.execute_input":"2024-07-28T12:57:50.593019Z","iopub.status.idle":"2024-07-28T12:57:51.679345Z","shell.execute_reply.started":"2024-07-28T12:57:50.592983Z","shell.execute_reply":"2024-07-28T12:57:51.678256Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os\nprint(os.listdir(\"../input/digit-recognizer\"))","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:51.681375Z","iopub.execute_input":"2024-07-28T12:57:51.681849Z","iopub.status.idle":"2024-07-28T12:57:51.688348Z","shell.execute_reply.started":"2024-07-28T12:57:51.681793Z","shell.execute_reply":"2024-07-28T12:57:51.687070Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport torch\nimport torch.nn as nn\nfrom torch.autograd import Variable\nfrom sklearn.model_selection import train_test_split\nfrom torch.utils.data import DataLoader, TensorDataset","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:51.689757Z","iopub.execute_input":"2024-07-28T12:57:51.690138Z","iopub.status.idle":"2024-07-28T12:57:55.850340Z","shell.execute_reply.started":"2024-07-28T12:57:51.690103Z","shell.execute_reply":"2024-07-28T12:57:55.849294Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train = pd.read_csv(r\"../input/digit-recognizer/train.csv\", dtype = np.float32)\n\ntargets_numpy = train.label.values\nfeatures_numpy = train.loc[:, train.columns != \"label\"]\nfeatures_numpy = features_numpy.values/255\n\nfeatures_train, features_test, targets_train, targets_test = train_test_split(features_numpy, \n                                                                              targets_numpy,\n                                                                              test_size = 0.2,\n                                                                              random_state=42)","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:55.852588Z","iopub.execute_input":"2024-07-28T12:57:55.853065Z","iopub.status.idle":"2024-07-28T12:57:59.899071Z","shell.execute_reply.started":"2024-07-28T12:57:55.853034Z","shell.execute_reply":"2024-07-28T12:57:59.898121Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# creating the tensors\nfeaturesTrain = torch.from_numpy(features_train)\nfeaturesTest = torch.from_numpy(features_test)\ntargetsTrain = torch.from_numpy(targets_train).type(torch.LongTensor)\ntargetsTest = torch.from_numpy(targets_test).type(torch.LongTensor)","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:59.900223Z","iopub.execute_input":"2024-07-28T12:57:59.900515Z","iopub.status.idle":"2024-07-28T12:57:59.935672Z","shell.execute_reply.started":"2024-07-28T12:57:59.900490Z","shell.execute_reply":"2024-07-28T12:57:59.934451Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"batch_size = 100\nn_iters = 10000\nnum_epochs = int(n_iters/(len(features_train) / batch_size))\nnum_epochs = int(num_epochs)\nnum_epochs","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:59.937054Z","iopub.execute_input":"2024-07-28T12:57:59.937822Z","iopub.status.idle":"2024-07-28T12:57:59.946103Z","shell.execute_reply.started":"2024-07-28T12:57:59.937787Z","shell.execute_reply":"2024-07-28T12:57:59.945028Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train = TensorDataset(featuresTrain, targetsTrain)\ntest = TensorDataset(featuresTest, targetsTest)","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:59.947638Z","iopub.execute_input":"2024-07-28T12:57:59.948411Z","iopub.status.idle":"2024-07-28T12:57:59.955556Z","shell.execute_reply.started":"2024-07-28T12:57:59.948373Z","shell.execute_reply":"2024-07-28T12:57:59.954493Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# data loader\ntrain_loader = DataLoader(train, batch_size = batch_size, shuffle=False)\ntest_loader = DataLoader(test, batch_size = batch_size, shuffle=False)","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:59.956941Z","iopub.execute_input":"2024-07-28T12:57:59.957278Z","iopub.status.idle":"2024-07-28T12:57:59.965669Z","shell.execute_reply.started":"2024-07-28T12:57:59.957250Z","shell.execute_reply":"2024-07-28T12:57:59.964679Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_loader","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:59.966930Z","iopub.execute_input":"2024-07-28T12:57:59.967372Z","iopub.status.idle":"2024-07-28T12:57:59.977338Z","shell.execute_reply.started":"2024-07-28T12:57:59.967301Z","shell.execute_reply":"2024-07-28T12:57:59.976254Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.imshow(features_numpy[12].reshape(28,28))\n# plt.axis(\"off\")\nplt.title(str(targets_numpy[12]))\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:57:59.980614Z","iopub.execute_input":"2024-07-28T12:57:59.980945Z","iopub.status.idle":"2024-07-28T12:58:00.243509Z","shell.execute_reply.started":"2024-07-28T12:57:59.980916Z","shell.execute_reply":"2024-07-28T12:58:00.242054Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# creating the rnn model\nclass RNNModel(nn.Module):\n    def __init__(self, input_dim, hidden_dim, layer_dim, output_dim):\n        super(RNNModel, self).__init__()\n        \n        # number of hidden dimensions\n        self.hidden_dim = hidden_dim\n        \n        #no. of layer dimension\n        self.layer_dim = layer_dim\n        \n        # RNN model\n        self.rnn = nn.RNN(input_dim, \n                          hidden_dim, \n                          layer_dim, \n                          batch_first=True, \n                          nonlinearity='relu')\n        \n        # readout layer\n        self.fc = nn.Linear(hidden_dim, output_dim)\n        \n    def forward(self, x):\n        # init initial hidden stage with 0\n        h0 = Variable(torch.zeros(self.layer_dim, x.size(0), self.hidden_dim))\n        \n        # one time step\n        out, hn = self.rnn(x,h0)\n        out = self.fc(out[:, -1, :])\n        \n        return out\n    \n    \n# crate RNN\ninput_dim = 28\nhidden_dim = 100\nlayer_dim = 2\noutput_dim = 10\n\nmodel = RNNModel(input_dim, hidden_dim, layer_dim, output_dim)\n\n# loss function\nerror = nn.CrossEntropyLoss()\n\n# SGD optimizer\nlearning_rate = 0.05\noptimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)\n        ","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:58:00.244808Z","iopub.execute_input":"2024-07-28T12:58:00.245286Z","iopub.status.idle":"2024-07-28T12:58:01.742094Z","shell.execute_reply.started":"2024-07-28T12:58:00.245251Z","shell.execute_reply":"2024-07-28T12:58:01.741051Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"\n# a = Variable(torch.zeros(1, 28, 1))\n# print(a)\n","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:58:01.743268Z","iopub.execute_input":"2024-07-28T12:58:01.743775Z","iopub.status.idle":"2024-07-28T12:58:01.749100Z","shell.execute_reply.started":"2024-07-28T12:58:01.743743Z","shell.execute_reply":"2024-07-28T12:58:01.747792Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"seq_dim = 28\nloss_list = []\niteration_list = []\naccuracy_list = []\ncount = 0\n\nfor epoch in range(num_epochs):\n    for i, (images,labels) in enumerate(train_loader):\n        \n        train = Variable(images.view(-1, seq_dim, input_dim))\n        labels = Variable(labels)\n        \n        #clear gradient\n        optimizer.zero_grad()\n        \n        # forward propagation\n        outputs = model(train)\n        \n        \n        # calaulate softmax and cross-entropy loss\n        loss = error(outputs, labels)\n        \n        # calculating gradient \n        loss.backward()\n        \n        # update parameters\n        optimizer.step()\n        \n        count += 1\n        \n        \n        if count%250 == 0:\n            # calculate accuracy\n            correct = 0\n            total = 0\n            \n            # iterate through dataset\n            for images, labels in test_loader:\n                images = Variable(images.view(-1, seq_dim, input_dim))\n                \n                #forward propagation\n                outputs = model(images)\n                \n                #get prediction from the maximum value\n                predicted = torch.max(outputs.data, 1)[1]\n                \n                # total number of labels\n                total += labels.size(0)\n                \n                correct += (predicted == labels).sum()\n                \n            accuracy = 100 * correct / float(total)\n            \n            # store loss and iteration \n            loss_list.append(loss.data)\n            iteration_list.append(count)\n            accuracy_list.append(accuracy)\n            \n            if count%500 == 0:\n                print('Iteration: {} Loss: {} Accuracy: {} %'.format(count, loss.data, accuracy))\n            \n            ","metadata":{"execution":{"iopub.status.busy":"2024-07-28T12:58:01.750597Z","iopub.execute_input":"2024-07-28T12:58:01.750972Z","iopub.status.idle":"2024-07-28T13:01:14.463354Z","shell.execute_reply.started":"2024-07-28T12:58:01.750936Z","shell.execute_reply":"2024-07-28T13:01:14.462162Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# visualization loss \nplt.plot(iteration_list,loss_list)\nplt.xlabel(\"Number of iteration\")\nplt.ylabel(\"Loss\")\nplt.title(\"RNN: Loss vs Number of iteration\")\nplt.show()\n\n# visualization accuracy \nplt.plot(iteration_list,accuracy_list,color = \"red\")\nplt.xlabel(\"Number of iteration\")\nplt.ylabel(\"Accuracy\")\nplt.title(\"RNN: Accuracy vs Number of iteration\")\nplt.savefig('graph.png')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-28T13:01:14.464638Z","iopub.execute_input":"2024-07-28T13:01:14.464959Z","iopub.status.idle":"2024-07-28T13:01:15.019516Z","shell.execute_reply.started":"2024-07-28T13:01:14.464931Z","shell.execute_reply":"2024-07-28T13:01:15.018250Z"},"trusted":true},"execution_count":null,"outputs":[]}]}