{"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_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np \nimport pandas as pd\nimport json\nfrom PIL import Image\nimport os\n\nimport torch\nimport torch.nn as nn\nimport torchvision\nimport torchvision.transforms as transforms\nfrom torch.utils.data import Dataset, DataLoader","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"BATCH = 32\nEPOCHS = 20\n\nLR = 0.003\n\nDEVICE = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n\nOUTPUT_DIR = './'\nif not os.path.exists(OUTPUT_DIR):\n    os.makedirs(OUTPUT_DIR)\n\nTRAIN_DIR = '../input/cassava-leaf-disease-classification/train_images/'\nTEST_DIR = '../input/cassava-leaf-disease-classification/test_images/'","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"labels = json.load(open(\"../input/cassava-leaf-disease-classification/label_num_to_disease_map.json\"))\ntrain = pd.read_csv('../input/cassava-leaf-disease-classification/train.csv')\nsample = pd.read_csv('../input/cassava-leaf-disease-classification/sample_submission.csv')\n\nX, Y = train['image_id'].values, train['label'].values\nX_test = [name for name in (os.listdir(TEST_DIR))]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"VAL_PCT = 0.3  # для тестирования возьмем 30% наших данных\nval_size = int(len(X)*VAL_PCT)\nprint(val_size)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train = X[:-val_size]\nY_train = Y[:-val_size]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(len(X_train), len(X_test))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class TrainData(Dataset):\n    def __init__(self, Dir, FNames, Labels, Transform):\n        self.dir = Dir\n        self.fnames = FNames\n        self.transform = Transform\n        self.lbs = Labels\n        \n    def __len__(self):\n        return len(self.fnames)\n\n    def __getitem__(self, index):\n        x = Image.open(os.path.join(self.dir, self.fnames[index]))  \n        return self.transform(x), self.lbs[index] \n        \nclass TestData(Dataset):\n    def __init__(self, Dir, FNames, Transform):\n        self.dir = Dir\n        self.fnames = FNames\n        self.transform = Transform\n        \n    def __len__(self):\n        return len(self.fnames)\n\n    def __getitem__(self, index):\n        x = Image.open(os.path.join(self.dir, self.fnames[index]))     \n        return self.transform(x), self.fnames[index]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_transform = transforms.Compose(\n    [#transforms.RandomResizedCrop((100, 100), scale=(0.8, 1.0)),\n     transforms.CenterCrop(224),\n     transforms.RandomRotation(90),\n     transforms.RandomHorizontalFlip(p=0.5),\n     transforms.RandomVerticalFlip(p=0.5),\n     transforms.ColorJitter(brightness=0.3, contrast=0.3, saturation=0.1, hue=0),\n     transforms.ToTensor(),\n     transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))\n    ])\n\ntest_transform = transforms.Compose(\n    [transforms.Resize((100, 100)),\n     transforms.ToTensor(),\n     transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test = [name for name in (os.listdir(TEST_DIR))]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"trainset = TrainData(TRAIN_DIR, X_train, Y_train, train_transform)\ntrainloader = DataLoader(trainset, batch_size=BATCH, shuffle=True, num_workers=4)\n\ntestset = TestData(TEST_DIR, X_test, test_transform)\ntestloader = DataLoader(testset, batch_size=1, shuffle=False, num_workers=4)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#model = torchvision.models.resnet50(pretrained=True) # PRETRAINED\n#model = torchvision.models.resnet152(pretrained=True)\nmodel = torchvision.models.resnet152()\nmodel.fc = nn.Linear(2048, 5, bias=True)\nmodel = model.to(DEVICE)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"criterion = nn.CrossEntropyLoss()\noptimizer = torch.optim.Adam(model.parameters(), lr=LR)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for epoch in range(EPOCHS):\n    tr_loss = 0.0\n\n    model = model.train()\n\n    for i, (images, labels) in enumerate(trainloader):\n        \n        images = images.to(DEVICE)\n        labels = labels.to(DEVICE)\n\n        logits = model(images)\n        loss = criterion(logits, labels)\n        optimizer.zero_grad()\n        loss.backward()\n        optimizer.step()\n\n        tr_loss += loss.detach().item()\n    \n    model.eval()\n    print('Epoch: %d | Loss: %.4f'%(epoch, tr_loss / i))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"s_ls = []\n\nwith torch.no_grad():\n    model.eval()\n    for image, fname in testloader: \n        image = image.to(DEVICE)\n        \n        logits = model(image)        \n        ps = torch.exp(logits)        \n        _, top_class = ps.topk(1, dim=1)\n        \n        for pred in top_class:\n            s_ls.append([fname[0], pred.item()])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sub = pd.DataFrame.from_records(s_ls, columns=['image_id', 'label'])\nsub.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sub.to_csv(\"submission.csv\", index=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}