{"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":"![title.jpg](attachment:d4f5691e-771e-455f-b172-83e29b1fd5d7.jpg)","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19"},"attachments":{"d4f5691e-771e-455f-b172-83e29b1fd5d7.jpg":{"image/jpeg":"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"}}},{"cell_type":"markdown","source":"<div style=\"text-align: center;\">\nVersion Control Log\n</div>\n\n|  Version  |  Date  |  Description of change  |\n| ---- | ---- |--------|\n|  2  |  27/11/2021  | Initial release|\n|  3  |  28/11/2021  | Add weights for d0~d7 and table and correct typo|\n|  8  |  29/11/2021  | Add log plot|","metadata":{}},{"cell_type":"markdown","source":"# Message\nThis is my first public notebook for me.  \nSo if you find the notebook is something wrong or strange, please tell me!  \nTo create the notebook, I spare the much time😅  \nI work for it from the evening to the midnight😪 So If you find the notebook is useful please upvote👍  \nThen, I will be very HAPPY😆  \nLet's fight together for helping our Great Barrier Reef💪  ","metadata":{}},{"cell_type":"markdown","source":"# Other Resources\nI also prepare [the discussion](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/290992) about EfficientDet.  \nSo please check it out.\nAlso **inference** notebook is under preparation.","metadata":{}},{"cell_type":"markdown","source":"# Install","metadata":{}},{"cell_type":"code","source":"!pip install --no-deps '../input/timm-package/timm-0.4.12-py3-none-any.whl' > /dev/null\n!pip install --no-deps '../input/pycocotools/pycocotools-2.0-cp37-cp37m-linux_x86_64.whl' > /dev/null","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Imports","metadata":{}},{"cell_type":"code","source":"import sys\nsys.path.insert(0, \"../input/timm-efficientdet-pytorch-fix-v3/timm_efficientdet_pytorch_fix_v3\")\nsys.path.insert(0, \"../input/omegaconf\")\nsys.path.insert(0, \"../input/albumentations-fix-v1/albumentations_fix_v1\")\n\nimport torch\nimport os\nfrom datetime import datetime\nimport time\nimport random\nimport cv2\nimport pandas as pd\nimport numpy as np\nimport albumentations as A\nimport matplotlib.pyplot as plt\nfrom albumentations.pytorch.transforms import ToTensorV2\nfrom sklearn.model_selection import StratifiedKFold\nfrom torch.utils.data import Dataset,DataLoader\nfrom torch.utils.data.sampler import SequentialSampler, RandomSampler\nfrom glob import glob\n\nSEED = 42\n\ndef seed_everything(seed):\n    random.seed(seed)\n    os.environ['PYTHONHASHSEED'] = str(seed)\n    np.random.seed(seed)\n    torch.manual_seed(seed)\n    torch.cuda.manual_seed(seed)\n    torch.backends.cudnn.deterministic = True\n    torch.backends.cudnn.benchmark = True\n\nseed_everything(SEED)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data_df = pd.read_csv('../input/tensorflow-great-barrier-reef/train.csv')\ndata_df = data_df[data_df.annotations != '[]'].reset_index()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cols = ['image_id','video_id', 'x', 'y', 'w', 'h']\nnew_data_df = pd.DataFrame(index=[], columns=cols)\n\nfor index, row in data_df.iterrows():\n    annotations = eval(row['annotations'])\n    for annotation in annotations:\n        tmp_row = pd.Series({\"image_id\":row[\"image_id\"],\"video_id\":row[\"video_id\"], \"x\":annotation[\"x\"], \"y\":annotation[\"y\"],\"w\":annotation[\"width\"],\"h\":annotation[\"height\"]})\n        new_data_df = new_data_df.append(tmp_row, ignore_index=True)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data_df = new_data_df","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"skf = StratifiedKFold(n_splits=3, shuffle=True, random_state=42)\ndf_folds = data_df[['image_id']].copy()\ndf_folds.loc[:, 'bbox_count'] = 1\ndf_folds = df_folds.groupby('image_id').count()\ndf_folds.loc[:, 'video_id'] = data_df[['image_id', 'video_id']].groupby('image_id').min()['video_id']\ndf_folds.loc[:, 'stratify_group'] = np.char.add(\n    df_folds['video_id'].values.astype(str),\n    df_folds['bbox_count'].apply(lambda x: f'_{x // 15}').values.astype(str)\n)\ndf_folds.loc[:, 'fold'] = 0\nfor fold_number, (train_index, val_index) in enumerate(skf.split(X=df_folds.index, y=df_folds['stratify_group'])):\n    df_folds.loc[df_folds.iloc[val_index].index, 'fold'] = fold_number","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def get_train_transforms():\n    return A.Compose(\n        [\n            A.RandomSizedCrop(min_max_height=(600, 600), height=720, width=1280, p=0.5),\n            A.OneOf([\n                A.HueSaturationValue(hue_shift_limit=0.2, sat_shift_limit= 0.2, \n                                     val_shift_limit=0.2, p=0.9),\n                A.RandomBrightnessContrast(brightness_limit=0.2, \n                                           contrast_limit=0.2, p=0.9),\n            ],p=0.9),\n            A.ToGray(p=0.01),\n            A.HorizontalFlip(p=0.5),\n            A.VerticalFlip(p=0.5),\n            A.Resize(height=512, width=512, p=1),\n            A.Cutout(num_holes=8, max_h_size=64, max_w_size=64, fill_value=0, p=0.5),\n            ToTensorV2(p=1.0),\n        ], \n        p=1.0, \n        bbox_params=A.BboxParams(\n            format='pascal_voc',\n            min_area=0, \n            min_visibility=0,\n            label_fields=['labels']\n        )\n    )\n\ndef get_valid_transforms():\n    return A.Compose(\n        [\n            A.Resize(height=512, width=512, p=1.0),\n            ToTensorV2(p=1.0),\n        ], \n        p=1.0, \n        bbox_params=A.BboxParams(\n            format='pascal_voc',\n            min_area=0, \n            min_visibility=0,\n            label_fields=['labels']\n        )\n    )","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Datasets","metadata":{}},{"cell_type":"code","source":"TRAIN_ROOT_PATH = '../input/tensorflow-great-barrier-reef/train_images'\n\nclass DatasetRetriever(Dataset):\n\n    def __init__(self, marking, image_ids, transforms=None, test=False):\n        super().__init__()\n\n        self.image_ids = image_ids\n        self.marking = marking\n        self.transforms = transforms\n        self.test = test\n\n    def __getitem__(self, index: int):\n        video_id_image_id = self.image_ids[index]\n        video_id_image_ids = video_id_image_id.split('-')\n        video_id = video_id_image_ids[0]\n        image_id = video_id_image_ids[1]\n        \n        image, boxes = self.load_image_and_boxes(index)\n\n        # there is only one class\n        labels = torch.ones((boxes.shape[0],), dtype=torch.int64)\n        \n        target = {}\n        target['boxes'] = boxes\n        target['labels'] = labels\n        target['image_id'] = torch.tensor([index])\n\n        if self.transforms:\n            for i in range(10):\n                sample = self.transforms(**{\n                    'image': image,\n                    'bboxes': target['boxes'],\n                    'labels': labels\n                })\n                if len(sample['bboxes']) > 0:\n                    image = sample['image']\n                    target['boxes'] = torch.stack(tuple(map(torch.tensor, zip(*sample['bboxes'])))).permute(1, 0)\n                    target['boxes'][:,[0,1,2,3]] = target['boxes'][:,[1,0,3,2]]  #yxyx: be warning\n                    break\n\n        return image, target, image_id, video_id\n\n    def __len__(self) -> int:\n        return self.image_ids.shape[0]\n\n    def load_image_and_boxes(self, index):\n        video_id_image_id = self.image_ids[index]\n        video_id_image_ids = video_id_image_id.split('-')\n        video_id = \"video_\" + video_id_image_ids[0]\n        image_id = video_id_image_ids[1]\n        \n        img_path = f'{TRAIN_ROOT_PATH}/{video_id}/{image_id}.jpg'\n        \n        assert os.path.isfile(img_path) == True\n        \n        image = cv2.imread(img_path, cv2.IMREAD_COLOR)\n        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB).astype(np.float32)\n        image /= 255.0\n        records = self.marking[self.marking['image_id'] == video_id_image_id]\n        boxes = records[['x', 'y', 'w', 'h']].values\n        boxes[:, 2] = boxes[:, 0] + boxes[:, 2]\n        boxes[:, 3] = boxes[:, 1] + boxes[:, 3]\n        return image, boxes","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fold_number = 0\n\ntrain_dataset = DatasetRetriever(\n    image_ids=df_folds[df_folds['fold'] != fold_number].index.values,\n    marking=data_df,\n    transforms=get_train_transforms(),\n    test=False,\n)\n\nvalidation_dataset = DatasetRetriever(\n    image_ids=df_folds[df_folds['fold'] == fold_number].index.values,\n    marking=data_df,\n    transforms=get_valid_transforms(),\n    test=True,\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Example","metadata":{}},{"cell_type":"code","source":"# If you find the image looks red, please run the cell again several times. \n# It is just caused by switching R and B chanell in visualization (maybe).\n# I will solve this bug in the next version.\nimage, target, image_id, video_id = train_dataset[0]\nboxes = target['boxes'].cpu().numpy().astype(np.int32)\nnumpy_image = image.permute(1,2,0).cpu().numpy()\nfig, ax = plt.subplots(1, 1, figsize=(16, 8))\n\nfor box in boxes:\n    cv2.rectangle(numpy_image, (box[1], box[0]), (box[3],  box[2]), (0, 1, 0), 2)\n    \nax.set_axis_off()\nax.imshow(numpy_image);","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class AverageMeter(object):\n    \"\"\"Computes and stores the average and current value\"\"\"\n    def __init__(self):\n        self.reset()\n\n    def reset(self):\n        self.val = 0\n        self.avg = 0\n        self.sum = 0\n        self.count = 0\n\n    def update(self, val, n=1):\n        self.val = val\n        self.sum += val * n\n        self.count += n\n        self.avg = self.sum / self.count","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import warnings\n\nwarnings.filterwarnings(\"ignore\")\n\nclass Fitter:\n    \n    def __init__(self, model, device, config):\n        self.config = config\n        self.epoch = 0\n\n        self.base_dir = f'./{config.folder}'\n        if not os.path.exists(self.base_dir):\n            os.makedirs(self.base_dir)\n        \n        self.log_path = f'{self.base_dir}/log.csv'\n        self.best_summary_loss = 10**5\n\n        self.model = model\n        self.device = device\n\n        param_optimizer = list(self.model.named_parameters())\n        no_decay = ['bias', 'LayerNorm.bias', 'LayerNorm.weight']\n        optimizer_grouped_parameters = [\n            {'params': [p for n, p in param_optimizer if not any(nd in n for nd in no_decay)], 'weight_decay': 0.001},\n            {'params': [p for n, p in param_optimizer if any(nd in n for nd in no_decay)], 'weight_decay': 0.0}\n        ] \n\n        self.optimizer = torch.optim.AdamW(self.model.parameters(), lr=config.lr)\n        self.scheduler = config.SchedulerClass(self.optimizer, **config.scheduler_params)\n        print(f'Fitter prepared. Device is {self.device}')\n        self.log(f'status,epoch,loss,time')\n\n    def fit(self, train_loader, validation_loader):\n        for e in range(self.config.n_epochs):\n            if self.config.verbose:\n                lr = self.optimizer.param_groups[0]['lr']\n                timestamp = datetime.utcnow().isoformat()\n                print(f'\\n{timestamp}\\nLR: {lr}')\n\n            t = time.time()\n            summary_loss = self.train_one_epoch(train_loader)\n\n            self.log(f'train,{self.epoch},{summary_loss.avg:.5f},{(time.time() - t):.5f}')\n            self.save(f'{self.base_dir}/last-checkpoint.bin')\n\n            t = time.time()\n            summary_loss = self.validation(validation_loader)\n\n            self.log(f'val,{self.epoch},{summary_loss.avg:.5f},{(time.time() - t):.5f}')\n            if summary_loss.avg < self.best_summary_loss:\n                self.best_summary_loss = summary_loss.avg\n                self.model.eval()\n                self.save(f'{self.base_dir}/best-checkpoint-{str(self.epoch).zfill(3)}epoch.bin')\n                for path in sorted(glob(f'{self.base_dir}/best-checkpoint-*epoch.bin'))[:-3]:\n                    os.remove(path)\n\n            if self.config.validation_scheduler:\n                self.scheduler.step(metrics=summary_loss.avg)\n\n            self.epoch += 1\n\n    def validation(self, val_loader):\n        self.model.eval()\n        summary_loss = AverageMeter()\n        t = time.time()\n        for step, sample in enumerate(val_loader):\n            images = sample[0]\n            targets = sample[1]\n            if self.config.verbose:\n                if step % self.config.verbose_step == 0:\n                    print(\n                        f'Val Step {step}/{len(val_loader)}, ' + \\\n                        f'summary_loss: {summary_loss.avg:.5f}, ' + \\\n                        f'time: {(time.time() - t):.5f}', end='\\r'\n                    )\n            with torch.no_grad():\n                images = torch.stack(images)\n                batch_size = images.shape[0]\n                images = images.to(self.device).float()\n                boxes = [target['boxes'].to(self.device).float() for target in targets]\n                labels = [target['labels'].to(self.device).float() for target in targets]\n\n                loss, _, _ = self.model(images, boxes, labels)\n                summary_loss.update(loss.detach().item(), batch_size)\n\n        return summary_loss\n\n    def train_one_epoch(self, train_loader):\n        self.model.train()\n        summary_loss = AverageMeter()\n        t = time.time()\n            \n        # for step, (images, targets, image_ids) in enumerate(train_loader):\n        for step, sample in enumerate(train_loader):\n            images = sample[0]\n            targets = sample[1]\n            \n            if self.config.verbose:\n                if step % self.config.verbose_step == 0:\n                    print(\n                        f'Train Step {step}/{len(train_loader)}, ' + \\\n                        f'summary_loss: {summary_loss.avg:.5f}, ' + \\\n                        f'time: {(time.time() - t):.5f}', end='\\r'\n                    )\n            \n            images = torch.stack(images)\n            images = images.to(self.device).float()\n            batch_size = images.shape[0]\n            boxes = [target['boxes'].to(self.device).float() for target in targets]\n            labels = [target['labels'].to(self.device).float() for target in targets]\n\n            self.optimizer.zero_grad()\n            \n            loss, _, _ = self.model(images, boxes, labels)\n            \n            loss.backward()\n\n            summary_loss.update(loss.detach().item(), batch_size)\n\n            self.optimizer.step()\n\n            if self.config.step_scheduler:\n                self.scheduler.step()\n\n        return summary_loss\n    \n    def save(self, path):\n        self.model.eval()\n        torch.save({\n            'model_state_dict': self.model.model.state_dict(),\n            'optimizer_state_dict': self.optimizer.state_dict(),\n            'scheduler_state_dict': self.scheduler.state_dict(),\n            'best_summary_loss': self.best_summary_loss,\n            'epoch': self.epoch,\n        }, path)\n\n    def load(self, path):\n        checkpoint = torch.load(path)\n        self.model.model.load_state_dict(checkpoint['model_state_dict'])\n        self.optimizer.load_state_dict(checkpoint['optimizer_state_dict'])\n        self.scheduler.load_state_dict(checkpoint['scheduler_state_dict'])\n        self.best_summary_loss = checkpoint['best_summary_loss']\n        self.epoch = checkpoint['epoch'] + 1\n        \n    def log(self, message):\n        if self.config.verbose:\n            print(message)\n        with open(self.log_path, 'a+') as logger:\n            logger.write(f'{message}\\n')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class TrainGlobalConfig:\n    num_workers = 2\n    batch_size = 3 \n    n_epochs = 3#4\n    lr = 0.0002\n\n    folder = 'effdet'\n\n    # -------------------\n    verbose = True\n    verbose_step = 1\n    # -------------------\n\n    # --------------------\n    step_scheduler = False  # do scheduler.step after optimizer.step\n    validation_scheduler = True  # do scheduler.step after validation stage loss\n\n    \n    SchedulerClass = torch.optim.lr_scheduler.ReduceLROnPlateau\n    scheduler_params = dict(\n        mode='min',\n        factor=0.5,\n        patience=1,\n        verbose=False, \n        threshold=0.0001,\n        threshold_mode='abs',\n        cooldown=0, \n        min_lr=1e-8,\n        eps=1e-08\n    )\n    # --------------------","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def collate_fn(batch):\n    return tuple(zip(*batch))\n\ndef run_training():\n    device = torch.device('cuda:0')\n    net.to(device)\n\n    train_loader = torch.utils.data.DataLoader(\n        train_dataset,\n        batch_size=TrainGlobalConfig.batch_size,\n        sampler=RandomSampler(train_dataset),\n        pin_memory=False,\n        drop_last=True,\n        num_workers=TrainGlobalConfig.num_workers,\n        collate_fn=collate_fn,\n    )\n    val_loader = torch.utils.data.DataLoader(\n        validation_dataset, \n        batch_size=TrainGlobalConfig.batch_size,\n        num_workers=TrainGlobalConfig.num_workers,\n        shuffle=False,\n        sampler=SequentialSampler(validation_dataset),\n        pin_memory=False,\n        collate_fn=collate_fn,\n    )\n\n    fitter = Fitter(model=net, device=device, config=TrainGlobalConfig)\n    fitter.fit(train_loader, val_loader)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from effdet import get_efficientdet_config, EfficientDet, DetBenchTrain\nfrom effdet.efficientdet import HeadNet\n\ndef get_net():\n    config = get_efficientdet_config('tf_efficientdet_d0')\n    config.image_size = 512\n    config.norm_kwargs=dict(eps=.001, momentum=.01)\n\n    net = EfficientDet(config, pretrained_backbone=False)\n    checkpoint = torch.load('../input/efficientdet-init-weights/efficientdet_d0-d92fd44f.pth')\n\n    net.load_state_dict(checkpoint)\n    net.class_net = HeadNet(config, num_outputs=config.num_classes)\n\n    return DetBenchTrain(net, config)\n\nnet = get_net()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"run_training()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Training time,loss summary\nI tried one epoch training for d0 through d7.  \nHere is the summary of training time and loss.  \nIt depends on the situation. So please use it for rough indication.  \nI don't know comparing the loss between different models is meaningfull but I add the section for them.\n\n|  D#  |  Time of one epoch[sec] <br>total/train/val  |Loss of one epoch <br> train/val |\n| :----: | :----: |:----: |\n|  0  |  384/319/65 | 20491/3643 |\n|  1  |  418/351/67 | 22103/4575 |\n|  2  |  451/381/70 | 18487/2568 |\n|  3  |  536/461/75 | 13545/1556 |\n|  4  |  632/554/78 | 12094/1421 |\n|  5  | 739/661/78  | 7951/839 |\n|  6  |  779/691/88 | 7504/589 |\n|  7  | 763/676/87  | 12749/901 |\n\n<div style=\"text-align: center;\">\nNote. this data were created on version 0 on 28/11/2021.\n</div>","metadata":{}},{"cell_type":"markdown","source":"","metadata":{}},{"cell_type":"code","source":"num_d = np.array([0, 1, 2, 3, 4, 5, 6, 7])\ntime = np.array([384, 418, 451,536,632,739,779,763])\nlabel = [\"D0\", \"D1\", \"D2\", \"D3\", \"D4\", \"D5\", \"D6\", \"D7\"]\nplt.bar(num_d, time, tick_label=label, align=\"center\")\nplt.title(\"Relationship between D# and time for one epoch  \")\nplt.xlabel(\"D#\")\nplt.ylabel(\"Time [sec]\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Plot log","metadata":{}},{"cell_type":"code","source":"log_df = pd.read_csv('./effdet/log.csv')\ntrain_log_df = log_df[log_df[\"status\"]==\"train\"]\nval_log_df = log_df[log_df[\"status\"]==\"val\"]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_log_df.plot(x='epoch', y='loss')\nplt.ylabel(u'loss') \nplt.title(u'train loss', size=16)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"val_log_df.plot(x='epoch', y='loss')\nplt.ylabel(u'loss')\nplt.title(u'validation loss', size=16)","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}