{"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":"markdown","source":"
\n

[Pytorch] ArcFace Starter

\n
","metadata":{}},{"cell_type":"markdown","source":"![](https://media.istockphoto.com/illustrations/the-whale-is-blowing-illustration-id164494826?k=20&m=164494826&s=612x612&w=0&h=SGm8bwFqE7-h_ekqaXOVfIUIpKN8aW2AAMcFSbvpwYg=)","metadata":{}},{"cell_type":"markdown","source":"

📌 Starter Notebook:

https://www.kaggle.com/debarshichanda/pytorch-happywhale-siamese-starter

\n\nhttps://www.kaggle.com/debarshichanda/pytorch-arcface-gem-pooling-starter","metadata":{}},{"cell_type":"markdown","source":"#

Install Required Libraries

","metadata":{}},{"cell_type":"code","source":"!pip install timm\n!pip install --upgrade wandb","metadata":{"_kg_hide-output":true,"execution":{"iopub.status.busy":"2022-02-15T14:58:57.671912Z","iopub.execute_input":"2022-02-15T14:58:57.672441Z","iopub.status.idle":"2022-02-15T14:59:13.292286Z","shell.execute_reply.started":"2022-02-15T14:58:57.672344Z","shell.execute_reply":"2022-02-15T14:59:13.291444Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Import Required Libraries 📚

","metadata":{}},{"cell_type":"code","source":"import os\nimport gc\nimport cv2\nimport math\nimport copy\nimport time\nimport random\n\n# For data manipulation\nimport numpy as np\nimport pandas as pd\n\n# Pytorch Imports\nimport torch\nimport torch.nn as nn\nimport torch.optim as optim\nimport torch.nn.functional as F\nfrom torch.optim import lr_scheduler\nfrom torch.utils.data import Dataset, DataLoader\nfrom torch.cuda import amp\n\nfrom transformers import AdamW, get_cosine_schedule_with_warmup\n\n# Utils\nimport joblib\nfrom tqdm import tqdm\nfrom collections import defaultdict\n\n# Sklearn Imports\nfrom sklearn.preprocessing import LabelEncoder\nfrom sklearn.model_selection import StratifiedKFold\n\n# For Image Models\nimport timm\n\n# Albumentations for augmentations\nimport albumentations as A\nfrom albumentations.pytorch import ToTensorV2\n\n# For colored terminal text\nfrom colorama import Fore, Back, Style\nb_ = Fore.BLUE\nsr_ = Style.RESET_ALL\n\nimport warnings\nwarnings.filterwarnings(\"ignore\")\n\n# For descriptive error messages\nos.environ['CUDA_LAUNCH_BLOCKING'] = \"1\"","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:13.294372Z","iopub.execute_input":"2022-02-15T14:59:13.294638Z","iopub.status.idle":"2022-02-15T14:59:16.626724Z","shell.execute_reply.started":"2022-02-15T14:59:13.294599Z","shell.execute_reply":"2022-02-15T14:59:16.625919Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"\"Weights\n\n Weights & Biases (W&B) is a set of machine learning tools that helps you build better models faster. Kaggle competitions require fast-paced model development and evaluation. There are a lot of components: exploring the training data, training different models, combining trained models in different combinations (ensembling), and so on.\n\n> ⏳ Lots of components = Lots of places to go wrong = Lots of time spent debugging\n\nW&B can be useful for Kaggle competition with it's lightweight and interoperable tools:\n\n* Quickly track experiments,
\n* Version and iterate on datasets,
\n* Evaluate model performance,
\n* Reproduce models,
\n* Visualize results and spot regressions,
\n* Share findings with colleagues.\n\nTo learn more about Weights and Biases check out this kernel.","metadata":{}},{"cell_type":"code","source":"import wandb\n\ntry:\n from kaggle_secrets import UserSecretsClient\n user_secrets = UserSecretsClient()\n api_key = user_secrets.get_secret(\"wandb_api\")\n wandb.login(key=api_key)\n anony = None\nexcept:\n anony = \"must\"\n print('If you want to use your W&B account, go to Add-ons -> Secrets and provide your W&B access token. Use the Label name as wandb_api. \\nGet your W&B access token from here: https://wandb.ai/authorize')","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.628486Z","iopub.execute_input":"2022-02-15T14:59:16.628750Z","iopub.status.idle":"2022-02-15T14:59:16.722044Z","shell.execute_reply.started":"2022-02-15T14:59:16.628715Z","shell.execute_reply":"2022-02-15T14:59:16.721279Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Training Configuration ⚙️

","metadata":{}},{"cell_type":"markdown","source":"```\nmodel = create_model('swin_large_patch4_window7_224', pretrained=True)\nmodel = create_model('swin_large_patch4_window7_224_in22k', pretrained=True)\nmodel = create_model('swin_large_patch4_window12_384', pretrained=True)\nmodel = create_model('swin_large_patch4_window12_384_in22k', pretrained=True)\n```","metadata":{}},{"cell_type":"code","source":"CONFIG = {\"seed\": 402,\n \"epochs\": 4,\n \"img_size\": 384,\n \"model_name\": \"swin_large_patch4_window12_384_in22k\",\n \"num_classes\": 15587,\n \"train_batch_size\": 4,\n \"valid_batch_size\": 8,\n \"learning_rate\": 2e-5,\n \"scheduler\": 'CosineAnnealingLR',\n \"min_lr\": 1e-7,\n \"T_max\": 500,\n \"weight_decay\": 1e-6,\n \"n_fold\": 5,\n \"n_accumulate\": 1,\n \"device\": torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\"),\n # ArcFace Hyperparameters\n \"s\": 30.0, \n \"m\": 0.50,\n \"ls_eps\": 0.0,\n \"easy_margin\": False\n }","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.724643Z","iopub.execute_input":"2022-02-15T14:59:16.724995Z","iopub.status.idle":"2022-02-15T14:59:16.764271Z","shell.execute_reply.started":"2022-02-15T14:59:16.724955Z","shell.execute_reply":"2022-02-15T14:59:16.763459Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Set Seed for Reproducibility

","metadata":{}},{"cell_type":"code","source":"def set_seed(seed=42):\n '''Sets the seed of the entire notebook so results are the same every time we run.\n This is for REPRODUCIBILITY.'''\n np.random.seed(seed)\n torch.manual_seed(seed)\n torch.cuda.manual_seed(seed)\n # When running on the CuDNN backend, two further options must be set\n torch.backends.cudnn.deterministic = True\n torch.backends.cudnn.benchmark = False\n # Set a fixed value for the hash seed\n os.environ['PYTHONHASHSEED'] = str(seed)\n \nset_seed(CONFIG['seed'])","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.765688Z","iopub.execute_input":"2022-02-15T14:59:16.765977Z","iopub.status.idle":"2022-02-15T14:59:16.776332Z","shell.execute_reply.started":"2022-02-15T14:59:16.765940Z","shell.execute_reply":"2022-02-15T14:59:16.775585Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ROOT_DIR = '../input/happy-whale-and-dolphin'\nTRAIN_DIR = '../input/happy-whale-and-dolphin/train_images'\nTEST_DIR = '../input/happy-whale-and-dolphin/test_images'","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.779845Z","iopub.execute_input":"2022-02-15T14:59:16.780062Z","iopub.status.idle":"2022-02-15T14:59:16.785195Z","shell.execute_reply.started":"2022-02-15T14:59:16.780032Z","shell.execute_reply":"2022-02-15T14:59:16.784238Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def get_train_file_path(id):\n return f\"{TRAIN_DIR}/{id}\"","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.786656Z","iopub.execute_input":"2022-02-15T14:59:16.786993Z","iopub.status.idle":"2022-02-15T14:59:16.793765Z","shell.execute_reply.started":"2022-02-15T14:59:16.786932Z","shell.execute_reply":"2022-02-15T14:59:16.793077Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Read the Data 📖

","metadata":{}},{"cell_type":"code","source":"df = pd.read_csv(f\"{ROOT_DIR}/train.csv\")\ndf['file_path'] = df['image'].apply(get_train_file_path)\ndf.head()","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.795191Z","iopub.execute_input":"2022-02-15T14:59:16.795492Z","iopub.status.idle":"2022-02-15T14:59:16.894497Z","shell.execute_reply.started":"2022-02-15T14:59:16.795454Z","shell.execute_reply":"2022-02-15T14:59:16.893694Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"encoder = LabelEncoder()\ndf['individual_id'] = encoder.fit_transform(df['individual_id'])\n\nwith open(\"le.pkl\", \"wb\") as fp:\n joblib.dump(encoder, fp)","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.895940Z","iopub.execute_input":"2022-02-15T14:59:16.896210Z","iopub.status.idle":"2022-02-15T14:59:16.930332Z","shell.execute_reply.started":"2022-02-15T14:59:16.896174Z","shell.execute_reply":"2022-02-15T14:59:16.929718Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Create Folds

","metadata":{}},{"cell_type":"code","source":"skf = StratifiedKFold(n_splits=CONFIG['n_fold'])\n\nfor fold, ( _, val_) in enumerate(skf.split(X=df, y=df.individual_id)):\n df.loc[val_ , \"kfold\"] = fold","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:16.931345Z","iopub.execute_input":"2022-02-15T14:59:16.931577Z","iopub.status.idle":"2022-02-15T14:59:17.393734Z","shell.execute_reply.started":"2022-02-15T14:59:16.931543Z","shell.execute_reply":"2022-02-15T14:59:17.393038Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Dataset Class

","metadata":{}},{"cell_type":"code","source":"class HappyWhaleDataset(Dataset):\n def __init__(self, df, transforms=None):\n self.df = df\n self.file_names = df['file_path'].values\n self.labels = df['individual_id'].values\n self.transforms = transforms\n \n def __len__(self):\n return len(self.df)\n \n def __getitem__(self, index):\n img_path = self.file_names[index]\n img = cv2.imread(img_path)\n img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n label = self.labels[index]\n \n if self.transforms:\n img = self.transforms(image=img)[\"image\"]\n \n return {\n 'image': img,\n 'label': torch.tensor(label, dtype=torch.long)\n }","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:17.395062Z","iopub.execute_input":"2022-02-15T14:59:17.395294Z","iopub.status.idle":"2022-02-15T14:59:17.403346Z","shell.execute_reply.started":"2022-02-15T14:59:17.395261Z","shell.execute_reply":"2022-02-15T14:59:17.402472Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Augmentations

","metadata":{}},{"cell_type":"code","source":"data_transforms = {\n \"train\": A.Compose([\n A.Resize(CONFIG['img_size'], CONFIG['img_size']),\n A.HorizontalFlip(p=0.5),\n A.VerticalFlip(p=0.5),\n A.Rotate(limit=30, p=0.5),\n A.Normalize(\n mean=[0.485, 0.456, 0.406], \n std=[0.229, 0.224, 0.225], \n max_pixel_value=255.0, \n p=1.0\n ),\n ToTensorV2()], p=1.),\n \n \"valid\": A.Compose([\n A.Resize(CONFIG['img_size'], CONFIG['img_size']),\n A.Normalize(\n mean=[0.485, 0.456, 0.406], \n std=[0.229, 0.224, 0.225], \n max_pixel_value=255.0, \n p=1.0\n ),\n ToTensorV2()], p=1.)\n}","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:17.405279Z","iopub.execute_input":"2022-02-15T14:59:17.405561Z","iopub.status.idle":"2022-02-15T14:59:17.416569Z","shell.execute_reply.started":"2022-02-15T14:59:17.405526Z","shell.execute_reply":"2022-02-15T14:59:17.415853Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

GeM Pooling

\n\nCode taken from GeM Pooling Explained\n\n![](https://i.imgur.com/thTgYWG.jpg)","metadata":{}},{"cell_type":"code","source":"class GeM(nn.Module):\n def __init__(self, p=3, eps=1e-6):\n super(GeM, self).__init__()\n self.p = nn.Parameter(torch.ones(1)*p)\n self.eps = eps\n\n def forward(self, x):\n return self.gem(x, p=self.p, eps=self.eps)\n \n def gem(self, x, p=3, eps=1e-6):\n return F.avg_pool2d(x.clamp(min=eps).pow(p), (x.size(-2), x.size(-1))).pow(1./p)\n \n def __repr__(self):\n return self.__class__.__name__ + \\\n '(' + 'p=' + '{:.4f}'.format(self.p.data.tolist()[0]) + \\\n ', ' + 'eps=' + str(self.eps) + ')'","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:17.420574Z","iopub.execute_input":"2022-02-15T14:59:17.421040Z","iopub.status.idle":"2022-02-15T14:59:17.430055Z","shell.execute_reply.started":"2022-02-15T14:59:17.421000Z","shell.execute_reply":"2022-02-15T14:59:17.429320Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

ArcFace

\n\nCode taken from Landmark2019-1st-and-3rd-Place-Solution","metadata":{}},{"cell_type":"code","source":"class ArcMarginProduct(nn.Module):\n r\"\"\"Implement of large margin arc distance: :\n Args:\n in_features: size of each input sample\n out_features: size of each output sample\n s: norm of input feature\n m: margin\n cos(theta + m)\n \"\"\"\n def __init__(self, in_features, out_features, s=30.0, \n m=0.50, easy_margin=False, ls_eps=0.0):\n super(ArcMarginProduct, self).__init__()\n self.in_features = in_features\n self.out_features = out_features\n self.s = s\n self.m = m\n self.ls_eps = ls_eps # label smoothing\n self.weight = nn.Parameter(torch.FloatTensor(out_features, in_features))\n nn.init.xavier_uniform_(self.weight)\n\n self.easy_margin = easy_margin\n self.cos_m = math.cos(m)\n self.sin_m = math.sin(m)\n self.th = math.cos(math.pi - m)\n self.mm = math.sin(math.pi - m) * m\n\n def forward(self, input, label):\n # --------------------------- cos(theta) & phi(theta) ---------------------\n cosine = F.linear(F.normalize(input), F.normalize(self.weight))\n sine = torch.sqrt(1.0 - torch.pow(cosine, 2))\n phi = cosine * self.cos_m - sine * self.sin_m\n if self.easy_margin:\n phi = torch.where(cosine > 0, phi, cosine)\n else:\n phi = torch.where(cosine > self.th, phi, cosine - self.mm)\n # --------------------------- convert label to one-hot ---------------------\n # one_hot = torch.zeros(cosine.size(), requires_grad=True, device='cuda')\n one_hot = torch.zeros(cosine.size(), device=CONFIG['device'])\n one_hot.scatter_(1, label.view(-1, 1).long(), 1)\n if self.ls_eps > 0:\n one_hot = (1 - self.ls_eps) * one_hot + self.ls_eps / self.out_features\n # -------------torch.where(out_i = {x_i if condition_i else y_i) ------------\n output = (one_hot * phi) + ((1.0 - one_hot) * cosine)\n output *= self.s\n\n return output","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:17.431220Z","iopub.execute_input":"2022-02-15T14:59:17.431610Z","iopub.status.idle":"2022-02-15T14:59:17.444867Z","shell.execute_reply.started":"2022-02-15T14:59:17.431575Z","shell.execute_reply":"2022-02-15T14:59:17.444165Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Create Model

","metadata":{}},{"cell_type":"code","source":"class HappyWhaleModel(nn.Module):\n def __init__(self, model_name, pretrained=True):\n super(HappyWhaleModel, self).__init__()\n self.model = timm.create_model(model_name, pretrained=pretrained)\n in_features = self.model.head.in_features\n self.model.head = nn.Identity()\n# self.model.avgpool = nn.Identity()\n# self.pooling = GeM()\n self.fc = ArcMarginProduct(in_features, \n CONFIG[\"num_classes\"],\n s=CONFIG[\"s\"], \n m=CONFIG[\"m\"], \n easy_margin=CONFIG[\"ls_eps\"], \n ls_eps=CONFIG[\"ls_eps\"])\n\n def forward(self, images, labels):\n features = self.model(images)\n# pooled_features = self.pooling(features).flatten(1)\n output = self.fc(features, labels)\n return output\n \nmodel = HappyWhaleModel(CONFIG['model_name'])\nmodel.to(CONFIG['device']);","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:17.445872Z","iopub.execute_input":"2022-02-15T14:59:17.447537Z","iopub.status.idle":"2022-02-15T14:59:25.534949Z","shell.execute_reply.started":"2022-02-15T14:59:17.447504Z","shell.execute_reply":"2022-02-15T14:59:25.534210Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Loss Function

","metadata":{}},{"cell_type":"code","source":"def criterion(outputs, labels):\n return nn.CrossEntropyLoss()(outputs, labels)","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.536238Z","iopub.execute_input":"2022-02-15T14:59:25.536491Z","iopub.status.idle":"2022-02-15T14:59:25.542099Z","shell.execute_reply.started":"2022-02-15T14:59:25.536457Z","shell.execute_reply":"2022-02-15T14:59:25.540007Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Training Function

","metadata":{}},{"cell_type":"code","source":"def train_one_epoch(model, optimizer, scheduler, dataloader, device, epoch):\n model.train()\n \n dataset_size = 0\n running_loss = 0.0\n \n bar = tqdm(enumerate(dataloader), total=len(dataloader))\n for step, data in bar:\n images = data['image'].to(device, dtype=torch.float)\n labels = data['label'].to(device, dtype=torch.long)\n \n batch_size = images.size(0)\n \n outputs = model(images, labels)\n loss = criterion(outputs, labels)\n loss = loss / CONFIG['n_accumulate']\n \n loss.backward()\n \n if (step + 1) % CONFIG['n_accumulate'] == 0:\n optimizer.step()\n\n # zero the parameter gradients\n optimizer.zero_grad()\n\n if scheduler is not None:\n scheduler.step()\n \n running_loss += (loss.item() * batch_size)\n dataset_size += batch_size\n \n epoch_loss = running_loss / dataset_size\n \n bar.set_postfix(Epoch=epoch, Train_Loss=epoch_loss,\n LR=optimizer.param_groups[0]['lr'])\n gc.collect()\n \n return epoch_loss","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.543477Z","iopub.execute_input":"2022-02-15T14:59:25.543728Z","iopub.status.idle":"2022-02-15T14:59:25.554208Z","shell.execute_reply.started":"2022-02-15T14:59:25.543692Z","shell.execute_reply":"2022-02-15T14:59:25.553198Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Validation Function

","metadata":{}},{"cell_type":"code","source":"@torch.inference_mode()\ndef valid_one_epoch(model, dataloader, device, epoch):\n model.eval()\n \n dataset_size = 0\n running_loss = 0.0\n \n bar = tqdm(enumerate(dataloader), total=len(dataloader))\n for step, data in bar: \n images = data['image'].to(device, dtype=torch.float)\n labels = data['label'].to(device, dtype=torch.long)\n \n batch_size = images.size(0)\n\n outputs = model(images, labels)\n loss = criterion(outputs, labels)\n \n running_loss += (loss.item() * batch_size)\n dataset_size += batch_size\n \n epoch_loss = running_loss / dataset_size\n \n bar.set_postfix(Epoch=epoch, Valid_Loss=epoch_loss,\n LR=optimizer.param_groups[0]['lr']) \n \n gc.collect()\n \n return epoch_loss","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.555389Z","iopub.execute_input":"2022-02-15T14:59:25.556112Z","iopub.status.idle":"2022-02-15T14:59:25.568556Z","shell.execute_reply.started":"2022-02-15T14:59:25.556073Z","shell.execute_reply":"2022-02-15T14:59:25.567688Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Run Training

","metadata":{}},{"cell_type":"code","source":"def run_training(model, optimizer, scheduler, device, num_epochs):\n # To automatically log gradients\n wandb.watch(model, log_freq=100)\n \n if torch.cuda.is_available():\n print(\"[INFO] Using GPU: {}\\n\".format(torch.cuda.get_device_name()))\n \n start = time.time()\n best_model_wts = copy.deepcopy(model.state_dict())\n best_epoch_loss = np.inf\n history = defaultdict(list)\n \n for epoch in range(1, num_epochs + 1): \n gc.collect()\n train_epoch_loss = train_one_epoch(model, optimizer, scheduler, \n dataloader=train_loader, \n device=CONFIG['device'], epoch=epoch)\n \n val_epoch_loss = valid_one_epoch(model, valid_loader, device=CONFIG['device'], \n epoch=epoch)\n \n history['Train Loss'].append(train_epoch_loss)\n history['Valid Loss'].append(val_epoch_loss)\n \n # Log the metrics\n wandb.log({\"Train Loss\": train_epoch_loss})\n wandb.log({\"Valid Loss\": val_epoch_loss})\n \n # deep copy the model\n if val_epoch_loss <= best_epoch_loss:\n print(f\"{b_}Validation Loss Improved ({best_epoch_loss} ---> {val_epoch_loss})\")\n best_epoch_loss = val_epoch_loss\n run.summary[\"Best Loss\"] = best_epoch_loss\n best_model_wts = copy.deepcopy(model.state_dict())\n PATH = \"Loss{:.4f}_epoch{:.0f}.bin\".format(best_epoch_loss, epoch)\n torch.save(model.state_dict(), PATH)\n # Save a model file from the current directory\n print(f\"Model Saved{sr_}\")\n \n print()\n \n end = time.time()\n time_elapsed = end - start\n print('Training complete in {:.0f}h {:.0f}m {:.0f}s'.format(\n time_elapsed // 3600, (time_elapsed % 3600) // 60, (time_elapsed % 3600) % 60))\n print(\"Best Loss: {:.4f}\".format(best_epoch_loss))\n \n # load best model weights\n model.load_state_dict(best_model_wts)\n \n return model, history","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.570166Z","iopub.execute_input":"2022-02-15T14:59:25.570760Z","iopub.status.idle":"2022-02-15T14:59:25.583127Z","shell.execute_reply.started":"2022-02-15T14:59:25.570671Z","shell.execute_reply":"2022-02-15T14:59:25.582367Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def fetch_scheduler(optimizer):\n if CONFIG['scheduler'] == 'CosineAnnealingLR':\n scheduler = lr_scheduler.CosineAnnealingLR(optimizer,T_max=CONFIG['T_max'], \n eta_min=CONFIG['min_lr'])\n elif CONFIG['scheduler'] == 'CosineAnnealingWarmRestarts':\n scheduler = lr_scheduler.CosineAnnealingWarmRestarts(optimizer,T_0=CONFIG['T_0'], \n eta_min=CONFIG['min_lr'])\n# elif CONFIG[\"scheduler\"] == \"\":\n# scheduler = get_cosine_schedule_with_warmup()\n \n elif CONFIG['scheduler'] == None:\n return None\n \n return scheduler","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.584569Z","iopub.execute_input":"2022-02-15T14:59:25.585101Z","iopub.status.idle":"2022-02-15T14:59:25.595208Z","shell.execute_reply.started":"2022-02-15T14:59:25.585064Z","shell.execute_reply":"2022-02-15T14:59:25.594525Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def prepare_loaders(df, fold):\n df_train = df[df.kfold != fold].reset_index(drop=True)\n df_valid = df[df.kfold == fold].reset_index(drop=True)\n \n train_dataset = HappyWhaleDataset(df_train, transforms=data_transforms[\"train\"])\n valid_dataset = HappyWhaleDataset(df_valid, transforms=data_transforms[\"valid\"])\n\n train_loader = DataLoader(train_dataset, batch_size=CONFIG['train_batch_size'], \n num_workers=2, shuffle=True, pin_memory=True, drop_last=True)\n valid_loader = DataLoader(valid_dataset, batch_size=CONFIG['valid_batch_size'], \n num_workers=2, shuffle=False, pin_memory=True)\n \n return train_loader, valid_loader","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.596677Z","iopub.execute_input":"2022-02-15T14:59:25.597233Z","iopub.status.idle":"2022-02-15T14:59:25.605089Z","shell.execute_reply.started":"2022-02-15T14:59:25.597196Z","shell.execute_reply":"2022-02-15T14:59:25.604323Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Prepare Dataloaders","metadata":{}},{"cell_type":"code","source":"train_loader, valid_loader = prepare_loaders(df, fold=0)","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.605968Z","iopub.execute_input":"2022-02-15T14:59:25.606484Z","iopub.status.idle":"2022-02-15T14:59:25.625251Z","shell.execute_reply.started":"2022-02-15T14:59:25.606447Z","shell.execute_reply":"2022-02-15T14:59:25.624572Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Define Optimizer and Scheduler","metadata":{}},{"cell_type":"code","source":"# optimizer = optim.Adam(model.parameters(), lr=CONFIG['learning_rate'], \n# weight_decay=CONFIG['weight_decay'])\n\nno_decay = [\"bias\", \"LayerNorm.weight\", \"LayerNorm.bias\"]\noptimizer_grouped_parameters = [\n {\n \"params\": [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)],\n \"weight_decay\": CONFIG['weight_decay'],\n },\n {\n \"params\": [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)],\n \"weight_decay\": 0.0,\n },\n]\noptimizer = AdamW(optimizer_grouped_parameters, lr=CONFIG['learning_rate'], weight_decay=CONFIG['weight_decay'])\n\nscheduler = fetch_scheduler(optimizer)","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.626316Z","iopub.execute_input":"2022-02-15T14:59:25.626702Z","iopub.status.idle":"2022-02-15T14:59:25.641013Z","shell.execute_reply.started":"2022-02-15T14:59:25.626666Z","shell.execute_reply":"2022-02-15T14:59:25.640240Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Start Training","metadata":{}},{"cell_type":"code","source":"run = wandb.init(project='HappyWhale', \n config=CONFIG,\n job_type='Train',\n tags=['arcface', 'gem-pooling', 'swin-transformer', '224'],\n anonymous='must')","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:25.642060Z","iopub.execute_input":"2022-02-15T14:59:25.642534Z","iopub.status.idle":"2022-02-15T14:59:32.858864Z","shell.execute_reply.started":"2022-02-15T14:59:25.642497Z","shell.execute_reply":"2022-02-15T14:59:32.857977Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model, history = run_training(model, optimizer, scheduler,\n device=CONFIG['device'],\n num_epochs=CONFIG['epochs'])","metadata":{"execution":{"iopub.status.busy":"2022-02-15T14:59:32.860611Z","iopub.execute_input":"2022-02-15T14:59:32.861037Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"run.finish()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#

Visualizations

\n\nView the Complete Dashboard Here ⮕","metadata":{}},{"cell_type":"markdown","source":"![](https://i.imgur.com/zD3rD0W.jpg)","metadata":{}},{"cell_type":"markdown","source":"![Upvote!](https://img.shields.io/badge/Upvote-If%20you%20like%20my%20work-07b3c8?style=for-the-badge&logo=kaggle)","metadata":{}}]}