diff --git a/pathology/tumor_detection/README.MD b/pathology/tumor_detection/README.MD index 7fe29da5ab..5d50e75a92 100644 --- a/pathology/tumor_detection/README.MD +++ b/pathology/tumor_detection/README.MD @@ -12,12 +12,11 @@ The model is based on ResNet18 with the last fully connected layer replaced by a All the data used to train and validate this model is from [Camelyon-16 Challenge](https://camelyon16.grand-challenge.org/). You can download all the images for "CAMELYON16" data set from various sources listed [here](https://camelyon17.grand-challenge.org/Data/). -Location information for training/validation patches (the location on the whole slide image where patches are extracted) are adopted from [NCRF/coords](https://github.com/baidu-research/NCRF/tree/master/coords). The reformatted coordinations and labels are stored in a json file (`dataset_0.json`), and can be downloaded from [here](https://drive.google.com/file/d/1m2pwko6hxwsxeDWZY2oSOV-_KT97Ol0o/view?usp=sharing) +Location information for training/validation patches (the location on the whole slide image where patches are extracted) are adopted from [NCRF/coords](https://github.com/baidu-research/NCRF/tree/master/coords). The reformatted coordinations and labels in CSV format for training (`training.csv`) can be found [here](https://drive.google.com/file/d/1httIjgji6U6rMIb0P8pE0F-hXFAuvQEf/view?usp=sharing) and for validation (`validation.csv`) can be found [here](https://drive.google.com/file/d/1tJulzl9m5LUm16IeFbOCoFnaSWoB6i5L/view?usp=sharing). -This pipeline expects the training/validation data (whole slide images) reside in `cfg["data_root"]/training/images`. By default `data_root` is pointing to `/workspace/data/medical/pathology/` You can easily modify it to point to a different directory by passing the following argument in the runtime: `--data-root /other/data/root/dir/`. +This pipeline expects the training/validation data (whole slide images) reside in `cfg["data_root"]/training/images`. By default `data_root` is pointing to the code folder `./`; however, you can easily modify it to point to a different directory by passing the following argument in the runtime: `--data-root /other/data/root/dir/`. -> `dataset_0_subset_0.json` is also provided [here](https://drive.google.com/file/d/1NCd0y4FR42maQpfZjzKlFSIX4oeKgysg/view?usp=sharing) to check the functionality of the pipeline using only two of the whole slide images: `tumor_001` and `tumor_101`.
-> This dataset should not be used for the real training or any perfomance evaluation. +> [`training_sub.csv`](https://drive.google.com/file/d/1rO8ZY-TrU9nrOsx-Udn1q5PmUYrLG3Mv/view?usp=sharing) and [`validation_sub.csv`](https://drive.google.com/file/d/130pqsrc2e9wiHIImL8w4fT_5NktEGel7/view?usp=sharing) is also provided to check the functionality of the pipeline using only two of the whole slide images: `tumor_001` (for training) and `tumor_101` (for validation). This dataset should not be used for the real training or any performance evaluation. ### Input and output formats diff --git a/pathology/tumor_detection/ignite/camelyon_train_evaluate.py b/pathology/tumor_detection/ignite/camelyon_train_evaluate.py index dd670f2f63..e3a74dc8e1 100644 --- a/pathology/tumor_detection/ignite/camelyon_train_evaluate.py +++ b/pathology/tumor_detection/ignite/camelyon_train_evaluate.py @@ -1,34 +1,17 @@ -import os - import logging +import os import time from argparse import ArgumentParser import numpy as np - +import pandas as pd import torch -from torch.optim import SGD, lr_scheduler - from ignite.metrics import Accuracy +from torch.optim import SGD, lr_scheduler import monai -from monai.data import DataLoader, load_decathlon_datalist -from monai.transforms import ( - ActivationsD, - AsDiscreteD, - CastToTypeD, - Compose, - RandFlipD, - RandRotate90D, - RandZoomD, - ScaleIntensityRangeD, - ToNumpyD, - TorchVisionD, - ToTensorD, -) -from monai.utils import first, set_determinism -from monai.optimizers import Novograd -from monai.engines import SupervisedTrainer, SupervisedEvaluator +from monai.data import DataLoader, PatchWSIDataset, CSVDataset +from monai.engines import SupervisedEvaluator, SupervisedTrainer from monai.handlers import ( CheckpointSaver, LrScheduleHandler, @@ -37,10 +20,24 @@ ValidationHandler, from_engine, ) - -from monai.apps.pathology.data import PatchWSIDataset from monai.networks.nets import TorchVisionFCModel - +from monai.optimizers import Novograd +from monai.transforms import ( + Activationsd, + AsDiscreted, + CastToTyped, + Compose, + GridSplitd, + Lambdad, + RandFlipd, + RandRotate90d, + RandZoomd, + ScaleIntensityRanged, + ToNumpyd, + TorchVisiond, + ToTensord, +) +from monai.utils import first, set_determinism torch.backends.cudnn.enabled = True set_determinism(seed=0, additional_settings=None) @@ -65,7 +62,7 @@ def set_device(cfg): if gpus and torch.cuda.is_available(): os.environ["CUDA_VISIBLE_DEVICES"] = ",".join([str(n) for n in gpus]) device = torch.device("cuda") - print(f'CUDA is being used with GPU ID(s): {os.environ["CUDA_VISIBLE_DEVICES"]}') + print(f'CUDA is being used with GPU Id(s): {os.environ["CUDA_VISIBLE_DEVICES"]}') else: device = torch.device("cpu") print("CPU only!") @@ -82,54 +79,66 @@ def train(cfg): # Build MONAI preprocessing train_preprocess = Compose( [ - ToTensorD(keys="image"), - TorchVisionD( + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + ToTensord(keys=("image")), + TorchVisiond( keys="image", name="ColorJitter", brightness=64.0 / 255.0, contrast=0.75, saturation=0.25, hue=0.04 ), - ToNumpyD(keys="image"), - RandFlipD(keys="image", prob=0.5), - RandRotate90D(keys="image", prob=0.5), - CastToTypeD(keys="image", dtype=np.float32), - RandZoomD(keys="image", prob=0.5, min_zoom=0.9, max_zoom=1.1), - ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), - ToTensorD(keys=("image", "label")), + ToNumpyd(keys="image"), + RandFlipd(keys="image", prob=0.5), + RandRotate90d(keys="image", prob=0.5, max_k=3, spatial_axes=(-2, -1)), + CastToTyped(keys="image", dtype=np.float32), + RandZoomd(keys="image", prob=0.5, min_zoom=0.9, max_zoom=1.1), + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensord(keys=("image", "label")), ] ) valid_preprocess = Compose( [ - CastToTypeD(keys="image", dtype=np.float32), - ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), - ToTensorD(keys=("image", "label")), + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + CastToTyped(keys="image", dtype=np.float32), + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensord(keys=("image", "label")), ] ) # __________________________________________________________________________ # Create MONAI dataset - train_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="training", - base_dir=cfg["data_root"], + train_data_list = CSVDataset( + cfg["train_file"], + col_groups={"image": 0, "patch_location": [2, 1], "label": [3, 6, 9, 4, 7, 10, 5, 8, 11]}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "training/images", x + ".tif")), ) - valid_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="validation", - base_dir=cfg["data_root"], + train_dataset = PatchWSIDataset( + data=train_data_list, + patch_size=cfg["region_size"], + patch_level=0, + transform=train_preprocess, + reader="openslide" if cfg["use_openslide"] else "cuCIM", ) - train_dataset = PatchWSIDataset( - train_json_info_list, - cfg["region_size"], - cfg["grid_shape"], - cfg["patch_size"], - train_preprocess, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", + valid_data_list = CSVDataset( + cfg["valid_file"], + col_groups={"image": 0, "patch_location": [2, 1], "label": [3, 6, 9, 4, 7, 10, 5, 8, 11]}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "training/images", x + ".tif")), ) valid_dataset = PatchWSIDataset( - valid_json_info_list, - cfg["region_size"], - cfg["grid_shape"], - cfg["patch_size"], - valid_preprocess, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", + data=valid_data_list, + patch_size=cfg["region_size"], + patch_level=0, + transform=valid_preprocess, + reader="openslide" if cfg["use_openslide"] else "cuCIM", ) # __________________________________________________________________________ @@ -141,12 +150,10 @@ def train(cfg): valid_dataset, num_workers=cfg["num_workers"], batch_size=cfg["batch_size"], pin_memory=True ) - # __________________________________________________________________________ - # Get sample batch and some info + # Check first sample first_sample = first(train_dataloader) if first_sample is None: - raise ValueError("Fist sample is None!") - + raise ValueError("First sample is None!") print("image: ") print(" shape", first_sample["image"].shape) print(" type: ", type(first_sample["image"])) @@ -194,9 +201,7 @@ def train(cfg): StatsHandler(output_transform=lambda x: None), TensorBoardStatsHandler(log_dir=log_dir, output_transform=lambda x: None), ] - val_postprocessing = Compose( - [ActivationsD(keys="pred", sigmoid=True), AsDiscreteD(keys="pred", threshold=0.5)] - ) + val_postprocessing = Compose([Activationsd(keys="pred", sigmoid=True), AsDiscreted(keys="pred", threshold=0.5)]) evaluator = SupervisedEvaluator( device=device, val_data_loader=valid_dataloader, @@ -219,9 +224,7 @@ def train(cfg): log_dir=cfg["logdir"], tag_name="train_loss", output_transform=from_engine(["loss"], first=True) ), ] - train_postprocessing = Compose( - [ActivationsD(keys="pred", sigmoid=True), AsDiscreteD(keys="pred", threshold=0.5)] - ) + train_postprocessing = Compose([Activationsd(keys="pred", sigmoid=True), AsDiscreted(keys="pred", threshold=0.5)]) trainer = SupervisedTrainer( device=device, @@ -241,24 +244,18 @@ def train(cfg): def main(): logging.basicConfig(level=logging.INFO) parser = ArgumentParser(description="Tumor detection on whole slide pathology images.") - parser.add_argument( - "--dataset", - type=str, - default="../dataset_0.json", - dest="dataset_json", - help="path to dataset json file", - ) parser.add_argument( "--root", type=str, - default="/workspace/data/medical/pathology/", - dest="data_root", - help="path to root folder of images containing training folder", + default="/workspace/data/medical/pathology", + help="path to image folder containing training/validation", ) + parser.add_argument("--train-file", type=str, default="training.csv", help="path to training data file") + parser.add_argument("--valid-file", type=str, default="validation.csv", help="path to training data file") parser.add_argument("--logdir", type=str, default="./logs/", dest="logdir", help="log directory") parser.add_argument("--rs", type=int, default=256 * 3, dest="region_size", help="region size") - parser.add_argument("--gs", type=int, default=3, dest="grid_shape", help="image grid shape (3x3)") + parser.add_argument("--gs", type=int, default=3, dest="grid_shape", help="image grid shape e.g 3 means 3x3") parser.add_argument("--ps", type=int, default=224, dest="patch_size", help="patch size") parser.add_argument("--bs", type=int, default=64, dest="batch_size", help="batch size") parser.add_argument("--ep", type=int, default=10, dest="n_epochs", help="number of epochs") diff --git a/pathology/tumor_detection/ignite/camelyon_train_evaluate_nvtx_profiling.py b/pathology/tumor_detection/ignite/camelyon_train_evaluate_nvtx_profiling.py index 8e62be68f5..581191d30f 100644 --- a/pathology/tumor_detection/ignite/camelyon_train_evaluate_nvtx_profiling.py +++ b/pathology/tumor_detection/ignite/camelyon_train_evaluate_nvtx_profiling.py @@ -3,41 +3,42 @@ import time from argparse import ArgumentParser -import monai import numpy as np +import pandas as pd import torch -from monai.apps.pathology.data import PatchWSIDataset -from monai.data import DataLoader, load_decathlon_datalist +from ignite.metrics import Accuracy +from torch.optim import SGD, lr_scheduler + +import monai +from monai.data import DataLoader, PatchWSIDataset, CSVDataset from monai.engines import SupervisedEvaluator, SupervisedTrainer from monai.handlers import ( CheckpointSaver, LrScheduleHandler, + RangeHandler, StatsHandler, TensorBoardStatsHandler, ValidationHandler, from_engine, ) -from monai.handlers.nvtx_handlers import RangeHandler from monai.networks.nets import TorchVisionFCModel from monai.optimizers import Novograd from monai.transforms import ( - ActivationsD, - AsDiscreteD, - CastToTypeD, + Activationsd, + AsDiscreted, + CastToTyped, Compose, - RandFlipD, - RandRotate90D, - RandZoomD, - ScaleIntensityRangeD, - ToNumpyD, - TorchVisionD, - ToTensorD, + GridSplitd, + Lambdad, + RandFlipd, + RandRotate90d, + RandZoomd, + ScaleIntensityRanged, + ToNumpyd, + TorchVisiond, + ToTensord, ) -from monai.transforms.nvtx import RangePopD, RangePushD -from monai.utils import Range, first, set_determinism -from torch.optim import SGD, lr_scheduler - -from ignite.metrics import Accuracy +from monai.utils import first, set_determinism, Range torch.backends.cudnn.enabled = True set_determinism(seed=0, additional_settings=None) @@ -62,7 +63,7 @@ def set_device(cfg): if gpus and torch.cuda.is_available(): os.environ["CUDA_VISIBLE_DEVICES"] = ",".join([str(n) for n in gpus]) device = torch.device("cuda") - print(f'CUDA is being used with GPU ID(s): {os.environ["CUDA_VISIBLE_DEVICES"]}') + print(f'CUDA is being used with GPU Id(s): {os.environ["CUDA_VISIBLE_DEVICES"]}') else: device = torch.device("cpu") print("CPU only!") @@ -79,85 +80,82 @@ def train(cfg): # Build MONAI preprocessing train_preprocess = Compose( [ - Range()(ToTensorD(keys="image")), - Range("ColorJitter")( - TorchVisionD( - keys="image", - name="ColorJitter", - brightness=64.0 / 255.0, - contrast=0.75, - saturation=0.25, - hue=0.04, - ) + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + ToTensord(keys=("image")), + TorchVisiond( + keys="image", name="ColorJitter", brightness=64.0 / 255.0, contrast=0.75, saturation=0.25, hue=0.04 ), - Range()(ToNumpyD(keys="image")), - Range()(RandFlipD(keys="image", prob=0.5)), - Range()(RandRotate90D(keys="image", prob=0.5)), - Range()(CastToTypeD(keys="image", dtype=np.float32)), - Range()(RandZoomD(keys="image", prob=0.5, min_zoom=0.9, max_zoom=1.1)), - Range()(ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0)), - ToTensorD(keys=("image", "label")), + ToNumpyd(keys="image"), + RandFlipd(keys="image", prob=0.5), + RandRotate90d(keys="image", prob=0.5, max_k=3, spatial_axes=(-2, -1)), + CastToTyped(keys="image", dtype=np.float32), + RandZoomd(keys="image", prob=0.5, min_zoom=0.9, max_zoom=1.1), + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensord(keys=("image", "label")), ] ) - train_preprocess = Range("Preprocessing")(train_preprocess) + train_preprocess = Range("Preprocessing", recursive=True)(train_preprocess) valid_preprocess = Compose( [ - CastToTypeD(keys="image", dtype=np.float32), - ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), - ToTensorD(keys=("image", "label")), + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + CastToTyped(keys="image", dtype=np.float32), + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensord(keys=("image", "label")), ] ) # __________________________________________________________________________ # Create MONAI dataset - train_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="training", - base_dir=cfg["data_root"], + train_data_list = CSVDataset( + cfg["train_file"], + col_groups={"image": 0, "patch_location": [2, 1], "label": [3, 6, 9, 4, 7, 10, 5, 8, 11]}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "training/images", x + ".tif")), ) - valid_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="validation", - base_dir=cfg["data_root"], + train_dataset = PatchWSIDataset( + data=train_data_list, + patch_size=cfg["region_size"], + patch_level=0, + transform=train_preprocess, + reader="openslide" if cfg["use_openslide"] else "cuCIM", ) - train_dataset = PatchWSIDataset( - train_json_info_list, - cfg["region_size"], - cfg["grid_shape"], - cfg["patch_size"], - train_preprocess, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", + valid_data_list = CSVDataset( + cfg["valid_file"], + col_groups={"image": 0, "patch_location": [2, 1], "label": [3, 6, 9, 4, 7, 10, 5, 8, 11]}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "training/images", x + ".tif")), ) valid_dataset = PatchWSIDataset( - valid_json_info_list, - cfg["region_size"], - cfg["grid_shape"], - cfg["patch_size"], - valid_preprocess, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", + data=valid_data_list, + patch_size=cfg["region_size"], + patch_level=0, + transform=valid_preprocess, + reader="openslide" if cfg["use_openslide"] else "cuCIM", ) # __________________________________________________________________________ # DataLoaders train_dataloader = DataLoader( - train_dataset, - num_workers=cfg["num_workers"], - batch_size=cfg["batch_size"], - pin_memory=True, + train_dataset, num_workers=cfg["num_workers"], batch_size=cfg["batch_size"], pin_memory=True ) valid_dataloader = DataLoader( - valid_dataset, - num_workers=cfg["num_workers"], - batch_size=cfg["batch_size"], - pin_memory=True, + valid_dataset, num_workers=cfg["num_workers"], batch_size=cfg["batch_size"], pin_memory=True ) - # __________________________________________________________________________ - # Get sample batch and some info + # Check first sample first_sample = first(train_dataloader) if first_sample is None: - raise ValueError("Fist sample is None!") - + raise ValueError("First sample is None!") print("image: ") print(" shape", first_sample["image"].shape) print(" type: ", type(first_sample["image"])) @@ -206,12 +204,7 @@ def train(cfg): StatsHandler(output_transform=lambda x: None), TensorBoardStatsHandler(log_dir=log_dir, output_transform=lambda x: None), ] - val_postprocessing = Compose( - [ - ActivationsD(keys="pred", sigmoid=True), - AsDiscreteD(keys="pred", threshold=0.5), - ] - ) + val_postprocessing = Compose([Activationsd(keys="pred", sigmoid=True), AsDiscreted(keys="pred", threshold=0.5)]) evaluator = SupervisedEvaluator( device=device, val_data_loader=valid_dataloader, @@ -228,27 +221,16 @@ def train(cfg): RangeHandler("Batch"), LrScheduleHandler(lr_scheduler=scheduler, print_lr=True), CheckpointSaver( - save_dir=cfg["logdir"], - save_dict={"net": model, "opt": optimizer}, - save_interval=1, - epoch_level=True, + save_dir=cfg["logdir"], save_dict={"net": model, "opt": optimizer}, save_interval=1, epoch_level=True ), StatsHandler(tag_name="train_loss", output_transform=from_engine(["loss"], first=True)), ValidationHandler(validator=evaluator, interval=1, epoch_level=True), TensorBoardStatsHandler( - log_dir=cfg["logdir"], - tag_name="train_loss", - output_transform=from_engine(["loss"], first=True), + log_dir=cfg["logdir"], tag_name="train_loss", output_transform=from_engine(["loss"], first=True) ), ] - train_postprocessing = Compose( - [ - RangePushD("Postprocessing"), - Range()(ActivationsD(keys="pred", sigmoid=True)), - Range()(AsDiscreteD(keys="pred", threshold=0.5)), - RangePopD(), - ] - ) + train_postprocessing = Compose([Activationsd(keys="pred", sigmoid=True), AsDiscreted(keys="pred", threshold=0.5)]) + train_postprocessing = Range("Postprocessing", recursive=True)(train_postprocessing) trainer = SupervisedTrainer( device=device, @@ -268,24 +250,18 @@ def train(cfg): def main(): logging.basicConfig(level=logging.INFO) parser = ArgumentParser(description="Tumor detection on whole slide pathology images.") - parser.add_argument( - "--dataset", - type=str, - default="../dataset_0.json", - dest="dataset_json", - help="path to dataset json file", - ) parser.add_argument( "--root", type=str, - default="/workspace/data/medical/pathology/", - dest="data_root", - help="path to root folder of images containing training folder", + default="/workspace/data/medical/pathology", + help="path to image folder containing training/validation", ) + parser.add_argument("--train-file", type=str, default="training.csv", help="path to training data file") + parser.add_argument("--valid-file", type=str, default="validation.csv", help="path to training data file") parser.add_argument("--logdir", type=str, default="./logs/", dest="logdir", help="log directory") parser.add_argument("--rs", type=int, default=256 * 3, dest="region_size", help="region size") - parser.add_argument("--gs", type=int, default=3, dest="grid_shape", help="image grid shape (3x3)") + parser.add_argument("--gs", type=int, default=3, dest="grid_shape", help="image grid shape e.g 3 means 3x3") parser.add_argument("--ps", type=int, default=224, dest="patch_size", help="patch size") parser.add_argument("--bs", type=int, default=64, dest="batch_size", help="batch size") parser.add_argument("--ep", type=int, default=10, dest="n_epochs", help="number of epochs") @@ -293,20 +269,10 @@ def main(): parser.add_argument("--openslide", action="store_true", dest="use_openslide", help="use OpenSlide") parser.add_argument("--no-amp", action="store_false", dest="amp", help="deactivate amp") - parser.add_argument( - "--no-novograd", - action="store_false", - dest="novograd", - help="deactivate novograd optimizer", - ) - parser.add_argument( - "--no-pretrain", - action="store_false", - dest="pretrain", - help="deactivate Imagenet weights", - ) + parser.add_argument("--no-novograd", action="store_false", dest="novograd", help="deactivate novograd optimizer") + parser.add_argument("--no-pretrain", action="store_false", dest="pretrain", help="deactivate Imagenet weights") - parser.add_argument("--cpu", type=int, default=0, dest="num_workers", help="number of workers") + parser.add_argument("--cpu", type=int, default=8, dest="num_workers", help="number of workers") parser.add_argument("--gpu", type=str, default="0", dest="gpu", help="which gpu to use") args = parser.parse_args() diff --git a/pathology/tumor_detection/ignite/profiling_camelyon_pipeline.ipynb b/pathology/tumor_detection/ignite/profiling_camelyon_pipeline.ipynb index 5a726239ee..657c98f9db 100644 --- a/pathology/tumor_detection/ignite/profiling_camelyon_pipeline.ipynb +++ b/pathology/tumor_detection/ignite/profiling_camelyon_pipeline.ipynb @@ -73,9 +73,9 @@ "source": [ "### Download data\n", "\n", - "The pipeline that we are profiling `camelyon_train_evaluate_nvtx_profiling.py` required [Camelyon-16 Challenge](https://camelyon16.grand-challenge.org/) dataset. You can download all the images for \"CAMELYON16\" data set from sources listed [here](https://camelyon17.grand-challenge.org/Data/), as well as the coordinations and labels (`dataset_0.json`), from [here](/view?usp=sharing)\n", + "The pipeline that we are profiling `camelyon_train_evaluate_nvtx_profiling.py` required [Camelyon-16 Challenge](https://camelyon16.grand-challenge.org/) dataset. You can download all the images for \"CAMELYON16\" data set from sources listed [here](https://camelyon17.grand-challenge.org/Data/). Also you can find the coordinations and labels for training (`training.csv`) [here](https://drive.google.com/file/d/1httIjgji6U6rMIb0P8pE0F-hXFAuvQEf/view?usp=sharing) and for validation (`validation.csv`) [here](https://drive.google.com/file/d/1tJulzl9m5LUm16IeFbOCoFnaSWoB6i5L/view?usp=sharing).\n", "\n", - "However, for the demo of this notebook, we are downloading a very small subset of Camelyon dataaset, which uses only one whole slide image `tumor_091.tif` .\n" + "However, for the demo of this notebook, we are downloading a very small subset of Camelyon dataset, which uses only one whole slide image `tumor_091.tif` .\n" ] }, { @@ -88,13 +88,13 @@ "output_type": "stream", "text": [ "Downloading...\n", - "From: https://drive.google.com/uc?id=1F-lR9tXoFkPkC1yueM-_TyaFk3CO7v0s\n", - "To: /home/bhashemian/workspace/tutorials/pathology/tumor_detection/ignite/dataset_0.json\n", - "100%|██████████| 1.10M/1.10M [00:00<00:00, 14.2MB/s]\n", + "From: https://drive.google.com/uc?id=1uWS4CXKD-NP_6-SgiQbQfhFMzbs0UJIr\n", + "To: /Users/bhashemian/workspace/tutorials/pathology/tumor_detection/ignite/training.csv\n", + "100%|██████████| 153k/153k [00:00<00:00, 1.91MB/s]\n", "Downloading...\n", "From: https://drive.google.com/uc?id=1OxAeCMVqH9FGpIWpAXSEJe6cLinEGQtF\n", - "To: /home/bhashemian/workspace/tutorials/pathology/tumor_detection/ignite/training/images/tumor_091.tif\n", - "546MB [00:05, 106MB/s]\n" + "To: /Users/bhashemian/workspace/tutorials/pathology/tumor_detection/ignite/training/images/tumor_091.tif\n", + "100%|██████████| 546M/546M [00:22<00:00, 24.1MB/s] \n" ] }, { @@ -109,9 +109,9 @@ } ], "source": [ - "# Download datset.json\n", - "dataset_url = \"https://drive.google.com/uc?id=1F-lR9tXoFkPkC1yueM-_TyaFk3CO7v0s\"\n", - "dataset_path = \"dataset_0.json\"\n", + "# Download training.csv\n", + "dataset_url = \"https://drive.google.com/uc?id=1uWS4CXKD-NP_6-SgiQbQfhFMzbs0UJIr\"\n", + "dataset_path = \"training.csv\"\n", "gdown.download(dataset_url, dataset_path, quiet=False)\n", "\n", "# Download images\n", @@ -132,74 +132,23 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Warning: LBR backtrace method is not supported on this platform. DWARF backtrace method will be used.\n", - "{'dataset_json': './dataset_0.json', 'data_root': './', 'logdir': './logs/', 'region_size': 768, 'grid_shape': 3, 'patch_size': 224, 'batch_size': 64, 'n_epochs': 10, 'lr': 0.001, 'use_openslide': False, 'amp': True, 'novograd': True, 'pretrain': True, 'num_workers': 0, 'gpu': '0'}\n", - "Logs and model are saved at './logs/220223-191443_resnet18_ps224_bs64_ep10_lr0.001'.\n", - "CUDA is being used with GPU ID(s): 0\n", - "[Plugin: cucim.kit.cuslide] Loading the dynamic library from: /opt/conda/lib/python3.8/site-packages/cucim/clara/cucim.kit.cuslide@21.10.01.so\n", - "Initializing plugin: cucim.kit.cuslide (interfaces: [cucim::io::IImageFormat v0.1]) (impl: cucim.kit.cuslide)\n", - "[Plugin: cucim.kit.cumed] Loading the dynamic library from: /opt/conda/lib/python3.8/site-packages/cucim/clara/cucim.kit.cumed@21.10.01.so\n", - "Initializing plugin: cucim.kit.cumed (interfaces: [cucim::io::IImageFormat v0.1]) (impl: cucim.kit.cumed)\n", - "image: \n", - " shape torch.Size([576, 3, 224, 224])\n", - " type: \n", - " dtype: torch.float32\n", - "labels: \n", - " shape torch.Size([576, 1, 1, 1])\n", - " type: \n", - " dtype: torch.float32\n", - "batch size: 64\n", - "train number of batches: 47\n", - "valid number of batches: 0\n", - "INFO:ignite.engine.engine.SupervisedTrainer:Engine run resuming from iteration 0, epoch 0 until 10 epochs\n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 1/47 -- train_loss: 0.6409 \n", - "Collecting data...\n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 2/47 -- train_loss: 0.7057 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 3/47 -- train_loss: 0.7131 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 4/47 -- train_loss: 0.6571 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 5/47 -- train_loss: 0.6917 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 6/47 -- train_loss: 0.6641 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 7/47 -- train_loss: 0.6660 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 8/47 -- train_loss: 0.6686 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 9/47 -- train_loss: 0.6390 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 10/47 -- train_loss: 0.6890 \n", - "Processing events...\n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 11/47 -- train_loss: 0.6213 \n", - "INFO:ignite.engine.engine.SupervisedTrainer:Epoch: 1/10, Iter: 12/47 -- train_loss: 0.6279 \n", - "Saving temporary \"/tmp/nsys-report-a4ba-75f3-bf1b-c100.qdstrm\" file to disk...\n", - "\n", - "\n", - "Sent signal 15 (Terminated) to target application's process group ID = 149165.\n", - "Use the kill option to modify this behavior.\n", - "Creating final output files...\n", - "Processing [1% ]\n", - "The target application terminated with signal 15 (SIGTERM)\n", - "Processing [===============================================================100%]\n", - "Saved report file to \"/tmp/nsys-report-a4ba-75f3-bf1b-c100.qdrep\"\n", - "Report file moved to \"/home/bhashemian/workspace/tutorials/pathology/tumor_detection/ignite/profile_report.qdrep\"\n", - "\n" - ] - } - ], + "outputs": [], "source": [ "!nsys profile \\\n", " --trace nvtx,osrt,cudnn,cuda, \\\n", " --delay 15 \\\n", - " --duration 45 \\\n", + " --duration 60 \\\n", " --show-output true \\\n", " --force-overwrite true \\\n", - " --output profile_report \\\n", + " --output profile_report.nsys-rep \\\n", " python camelyon_train_evaluate_nvtx_profiling.py \\\n", " --cpu 0 \\\n", - " --dataset ./dataset_0.json \\\n", - " --root ./" + " --train-file ./training.csv \\\n", + " --valid-file ./training.csv \\\n", + " --root ./ \\\n", + " --bs 10" ] }, { @@ -211,35 +160,38 @@ "name": "stdout", "output_type": "stream", "text": [ - "Generating SQLite file profile_report.sqlite from profile_report.qdrep\n", - "Exporting 450058 events: [=================================================100%]\n", + "Generating SQLite file profile_report.sqlite from profile_report.nsys-rep\n", + "Exporting 265495 events: [=================================================100%]\n", "Using profile_report.sqlite for SQL queries.\n", - "Running [/usr/local/cuda-11.5/NsightSystems-cli-2021.3.2/target-linux-x64/reports/nvtxppsum.py profile_report.sqlite]... \n", + "Running [/usr/local/cuda-11.6/NsightSystems-cli-2021.5.2/target-linux-x64/reports/nvtxppsum.py profile_report.sqlite]... \n", "\n", - "+---------+-----------------+-----------+--------------+--------------+--------------+--------------+----------------------+\n", - "| Time(%) | Total Time (ns) | Instances | Average (ns) | Minimum (ns) | Maximum (ns) | StdDev (ns) | Range |\n", - "+---------+-----------------+-----------+--------------+--------------+--------------+--------------+----------------------+\n", - "| 34.4 | 40036304081 | 9 | 4448478231.2 | 782631451 | 10036025423 | 2388320102.6 | Batch |\n", - "| 30.3 | 35191327503 | 525 | 67031100.0 | 39060250 | 5297236602 | 229442683.8 | Preprocessing |\n", - "| 26.7 | 31077190736 | 4724 | 6578575.5 | 2991869 | 5222838076 | 76115696.2 | ColorJitter |\n", - "| 4.0 | 4666811337 | 8 | 583351417.1 | 543904689 | 669988364 | 46262035.8 | Iteration |\n", - "| 1.3 | 1493595476 | 4716 | 316708.1 | 12442 | 4689389 | 547614.2 | RandZoomd |\n", - "| 1.1 | 1263019896 | 4716 | 267815.9 | 215967 | 692103 | 41615.4 | ScaleIntensityRanged |\n", - "| 0.6 | 650986980 | 5184 | 125576.2 | 114374 | 339141 | 11487.2 | Postprocessing |\n", - "| 0.3 | 391508855 | 5184 | 75522.5 | 68872 | 154239 | 6759.0 | AsDiscreted |\n", - "| 0.3 | 359299648 | 4716 | 76187.4 | 50648 | 235844 | 16864.6 | ToNumpyd |\n", - "| 0.3 | 299120139 | 4716 | 63426.7 | 33967 | 131942 | 8541.3 | CastToTyped |\n", - "| 0.2 | 258728731 | 4716 | 54861.9 | 8278 | 994322 | 56994.7 | RandRotate90d |\n", - "| 0.2 | 232628941 | 5184 | 44874.4 | 40170 | 172579 | 5427.0 | Activationsd |\n", - "| 0.2 | 175299329 | 4716 | 37171.2 | 2872 | 230035 | 34228.3 | RandFlipd |\n", - "| 0.1 | 62369507 | 8 | 7796188.4 | 7069163 | 10072025 | 1083591.7 | ResNet18 |\n", - "| 0.1 | 59502039 | 4725 | 12593.0 | 6503 | 932484 | 18170.6 | ToTensord |\n", - "| 0.0 | 1964335 | 8 | 245541.9 | 220073 | 305026 | 31930.6 | Loss |\n", - "+---------+-----------------+-----------+--------------+--------------+--------------+--------------+----------------------+\n", + "+----------+-----------------+-----------+--------------+--------------+------------+-------------+--------------+--------------------------+\n", + "| Time (%) | Total Time (ns) | Instances | Avg (ns) | Med (ns) | Min (ns) | Max (ns) | StdDev (ns) | Range |\n", + "+----------+-----------------+-----------+--------------+--------------+------------+-------------+--------------+--------------------------+\n", + "| 28.7 | 33706579200 | 5 | 6741315840.0 | 6324451800.0 | 176995000 | 13682363400 | 4889241407.0 | Iteration |\n", + "| 21.3 | 25011936300 | 5 | 5002387260.0 | 4787481900.0 | 2873517700 | 8072929200 | 2035498721.7 | Batch |\n", + "| 20.3 | 23839370600 | 50 | 476787412.0 | 376589400.0 | 220633100 | 1154097400 | 276441893.0 | Preprocessing |\n", + "| 19.3 | 22742525900 | 450 | 50538946.4 | 36570950.0 | 18874200 | 202062000 | 36166736.1 | TorchVisiond_ColorJitter |\n", + "| 9.4 | 11044461700 | 5 | 2208892340.0 | 1996530400.0 | 148099300 | 4407900900 | 1534918799.1 | ResNet18 |\n", + "| 0.3 | 384269900 | 450 | 853933.1 | 65400.0 | 21000 | 22487800 | 2212634.3 | RandZoomd |\n", + "| 0.2 | 244892100 | 450 | 544204.7 | 441950.0 | 321800 | 8677700 | 541248.4 | ScaleIntensityRanged |\n", + "| 0.1 | 128083500 | 450 | 284630.0 | 243900.0 | 187400 | 4721600 | 230932.6 | Postprocessing |\n", + "| 0.1 | 91848800 | 450 | 204108.4 | 176450.0 | 128700 | 745700 | 87187.5 | ToNumpyd |\n", + "| 0.1 | 65417500 | 450 | 145372.2 | 117000.0 | 90200 | 4613300 | 219185.3 | AsDiscreted |\n", + "| 0.1 | 59017500 | 450 | 131150.0 | 82250.0 | 17400 | 1050600 | 123950.6 | RandRotate90d |\n", + "| 0.0 | 55917100 | 50 | 1118342.0 | 882450.0 | 685900 | 5798000 | 801880.3 | GridSplitd |\n", + "| 0.0 | 54120900 | 450 | 120268.7 | 100700.0 | 68500 | 721200 | 59828.4 | ToTensord |\n", + "| 0.0 | 51677300 | 450 | 114838.4 | 101350.0 | 67500 | 1154000 | 60160.7 | CastToTyped |\n", + "| 0.0 | 50674700 | 450 | 112610.4 | 95650.0 | 71300 | 613700 | 51643.2 | ToTensord_2 |\n", + "| 0.0 | 48966300 | 450 | 108814.0 | 95900.0 | 75200 | 428700 | 44791.2 | Activationsd |\n", + "| 0.0 | 39524100 | 450 | 87831.3 | 45950.0 | 7600 | 2748400 | 153536.8 | RandFlipd |\n", + "| 0.0 | 2460800 | 50 | 49216.0 | 39450.0 | 28700 | 146100 | 23219.6 | Lambdad |\n", + "| 0.0 | 1074200 | 4 | 268550.0 | 194150.0 | 142200 | 543700 | 186523.3 | Loss |\n", + "+----------+-----------------+-----------+--------------+--------------+------------+-------------+--------------+--------------------------+\n", "\n", - "Running [/usr/local/cuda-11.5/NsightSystems-cli-2021.3.2/target-linux-x64/reports/nvtxppsum.py profile_report.sqlite] to [profile_report_nvtxppsum.csv]... PROCESSED\n", + "Running [/usr/local/cuda-11.6/NsightSystems-cli-2021.5.2/target-linux-x64/reports/nvtxppsum.py profile_report.sqlite] to [profile_report_nvtxppsum.csv]... PROCESSED\n", "\n", - "Running [/usr/local/cuda-11.5/NsightSystems-cli-2021.3.2/target-linux-x64/reports/nvtxpptrace.py profile_report.sqlite] to [profile_report_nvtxpptrace.csv]... PROCESSED\n", + "Running [/usr/local/cuda-11.6/NsightSystems-cli-2021.5.2/target-linux-x64/reports/nvtxpptrace.py profile_report.sqlite] to [profile_report_nvtxpptrace.csv]... PROCESSED\n", "\n" ] } @@ -250,7 +202,7 @@ " --format table,csv \\\n", " --output -,. \\\n", " --force-overwrite true \\\n", - " profile_report.qdrep" + " profile_report.nsys-rep" ] }, { @@ -271,13 +223,16 @@ "# Ordered list of NVTX range for all training transforms\n", "transforms = [\n", " \"ToTensord\",\n", - " \"ColorJitter\",\n", + " \"Lambdad\",\n", + " \"GridSplitd\",\n", + " \"TorchVisiond_ColorJitter\",\n", " \"ToNumpyd\",\n", " \"RandFlipd\",\n", " \"RandRotate90d\",\n", " \"CastToTyped\",\n", " \"RandZoomd\",\n", " \"ScaleIntensityRanged\",\n", + " \"ToTensord_2\",\n", " \"Activationsd\",\n", " \"AsDiscreted\",\n", "]" @@ -316,10 +271,18 @@ " \n", " \n", " \n", - " Average\n", - " Minimum\n", - " Maximum\n", - " StdDev\n", + " Time (%)\n", + " Total Time (ns)\n", + " Instances\n", + " Avg (ns)\n", + " Med (ns)\n", + " Min (ns)\n", + " Max (ns)\n", + " StdDev (ns)\n", + " avg%\n", + " std%\n", + " min%\n", + " max%\n", " \n", " \n", " Range\n", @@ -327,96 +290,264 @@ " \n", " \n", " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", " ToTensord\n", - " 0.167288\n", - " 0.086387\n", - " 12.387309\n", - " 0.241382\n", + " 0.0\n", + " 54120900\n", + " 450\n", + " 120268.7\n", + " 100700.0\n", + " 68500\n", + " 721200\n", + " 59828.4\n", + " 0.222186\n", + " 0.110528\n", + " 0.126548\n", + " 1.332357\n", + " \n", + " \n", + " Lambdad\n", + " 0.0\n", + " 2460800\n", + " 50\n", + " 49216.0\n", + " 39450.0\n", + " 28700\n", + " 146100\n", + " 23219.6\n", + " 0.090922\n", + " 0.042896\n", + " 0.053021\n", + " 0.269908\n", " \n", " \n", - " ColorJitter\n", - " 87.391149\n", - " 39.744603\n", - " 69381.254333\n", - " 1011.136551\n", + " GridSplitd\n", + " 0.0\n", + " 55917100\n", + " 50\n", + " 1118342.0\n", + " 882450.0\n", + " 685900\n", + " 5798000\n", + " 801880.3\n", + " 2.066044\n", + " 1.481407\n", + " 1.267143\n", + " 10.711323\n", + " \n", + " \n", + " TorchVisiond_ColorJitter\n", + " 19.3\n", + " 22742525900\n", + " 450\n", + " 50538946.4\n", + " 36570950.0\n", + " 18874200\n", + " 202062000\n", + " 36166736.1\n", + " 93.366500\n", + " 66.815037\n", + " 34.868515\n", + " 373.292740\n", " \n", " \n", " ToNumpyd\n", - " 1.012089\n", - " 0.672818\n", - " 3.133000\n", - " 0.224033\n", + " 0.1\n", + " 91848800\n", + " 450\n", + " 204108.4\n", + " 176450.0\n", + " 128700\n", + " 745700\n", + " 87187.5\n", + " 0.377073\n", + " 0.161072\n", + " 0.237763\n", + " 1.377619\n", " \n", " \n", " RandFlipd\n", - " 0.493790\n", - " 0.038152\n", - " 3.055832\n", - " 0.454696\n", + " 0.0\n", + " 39524100\n", + " 450\n", + " 87831.3\n", + " 45950.0\n", + " 7600\n", + " 2748400\n", + " 153536.8\n", + " 0.162261\n", + " 0.283646\n", + " 0.014040\n", + " 5.077440\n", " \n", " \n", " RandRotate90d\n", - " 0.728797\n", - " 0.109967\n", - " 13.208778\n", - " 0.757129\n", + " 0.1\n", + " 59017500\n", + " 450\n", + " 131150.0\n", + " 82250.0\n", + " 17400\n", + " 1050600\n", + " 123950.6\n", + " 0.242289\n", + " 0.228988\n", + " 0.032145\n", + " 1.940896\n", " \n", " \n", " CastToTyped\n", - " 0.842573\n", - " 0.451225\n", - " 1.752745\n", - " 0.113464\n", + " 0.0\n", + " 51677300\n", + " 450\n", + " 114838.4\n", + " 101350.0\n", + " 67500\n", + " 1154000\n", + " 60160.7\n", + " 0.212154\n", + " 0.111142\n", + " 0.124701\n", + " 2.131919\n", " \n", " \n", " RandZoomd\n", - " 4.207215\n", - " 0.165282\n", - " 62.294807\n", - " 7.274620\n", + " 0.3\n", + " 384269900\n", + " 450\n", + " 853933.1\n", + " 65400.0\n", + " 21000\n", + " 22487800\n", + " 2212634.3\n", + " 1.577570\n", + " 4.087658\n", + " 0.038796\n", + " 41.544340\n", " \n", " \n", " ScaleIntensityRanged\n", - " 3.557721\n", - " 2.868950\n", - " 9.194038\n", - " 0.552828\n", + " 0.2\n", + " 244892100\n", + " 450\n", + " 544204.7\n", + " 441950.0\n", + " 321800\n", + " 8677700\n", + " 541248.4\n", + " 1.005373\n", + " 0.999911\n", + " 0.594499\n", + " 16.031329\n", + " \n", + " \n", + " ToTensord_2\n", + " 0.0\n", + " 50674700\n", + " 450\n", + " 112610.4\n", + " 95650.0\n", + " 71300\n", + " 613700\n", + " 51643.2\n", + " 0.208038\n", + " 0.095407\n", + " 0.131721\n", + " 1.133760\n", " \n", " \n", " Activationsd\n", - " 0.596121\n", - " 0.533627\n", - " 2.292575\n", - " 0.072093\n", + " 0.0\n", + " 48966300\n", + " 450\n", + " 108814.0\n", + " 95900.0\n", + " 75200\n", + " 428700\n", + " 44791.2\n", + " 0.201025\n", + " 0.082748\n", + " 0.138926\n", + " 0.791988\n", " \n", " \n", " AsDiscreted\n", - " 1.003256\n", - " 0.914910\n", - " 2.048943\n", - " 0.089788\n", + " 0.1\n", + " 65417500\n", + " 450\n", + " 145372.2\n", + " 117000.0\n", + " 90200\n", + " 4613300\n", + " 219185.3\n", + " 0.268563\n", + " 0.404927\n", + " 0.166637\n", + " 8.522688\n", " \n", " \n", "\n", "" ], "text/plain": [ - " Average Minimum Maximum StdDev\n", - "Range \n", - "ToTensord 0.167288 0.086387 12.387309 0.241382\n", - "ColorJitter 87.391149 39.744603 69381.254333 1011.136551\n", - "ToNumpyd 1.012089 0.672818 3.133000 0.224033\n", - "RandFlipd 0.493790 0.038152 3.055832 0.454696\n", - "RandRotate90d 0.728797 0.109967 13.208778 0.757129\n", - "CastToTyped 0.842573 0.451225 1.752745 0.113464\n", - "RandZoomd 4.207215 0.165282 62.294807 7.274620\n", - "ScaleIntensityRanged 3.557721 2.868950 9.194038 0.552828\n", - "Activationsd 0.596121 0.533627 2.292575 0.072093\n", - "AsDiscreted 1.003256 0.914910 2.048943 0.089788" + " Time (%) Total Time (ns) Instances Avg (ns) \\\n", + "Range \n", + "ToTensord 0.0 54120900 450 120268.7 \n", + "Lambdad 0.0 2460800 50 49216.0 \n", + "GridSplitd 0.0 55917100 50 1118342.0 \n", + "TorchVisiond_ColorJitter 19.3 22742525900 450 50538946.4 \n", + "ToNumpyd 0.1 91848800 450 204108.4 \n", + "RandFlipd 0.0 39524100 450 87831.3 \n", + "RandRotate90d 0.1 59017500 450 131150.0 \n", + "CastToTyped 0.0 51677300 450 114838.4 \n", + "RandZoomd 0.3 384269900 450 853933.1 \n", + "ScaleIntensityRanged 0.2 244892100 450 544204.7 \n", + "ToTensord_2 0.0 50674700 450 112610.4 \n", + "Activationsd 0.0 48966300 450 108814.0 \n", + "AsDiscreted 0.1 65417500 450 145372.2 \n", + "\n", + " Med (ns) Min (ns) Max (ns) StdDev (ns) \\\n", + "Range \n", + "ToTensord 100700.0 68500 721200 59828.4 \n", + "Lambdad 39450.0 28700 146100 23219.6 \n", + "GridSplitd 882450.0 685900 5798000 801880.3 \n", + "TorchVisiond_ColorJitter 36570950.0 18874200 202062000 36166736.1 \n", + "ToNumpyd 176450.0 128700 745700 87187.5 \n", + "RandFlipd 45950.0 7600 2748400 153536.8 \n", + "RandRotate90d 82250.0 17400 1050600 123950.6 \n", + "CastToTyped 101350.0 67500 1154000 60160.7 \n", + "RandZoomd 65400.0 21000 22487800 2212634.3 \n", + "ScaleIntensityRanged 441950.0 321800 8677700 541248.4 \n", + "ToTensord_2 95650.0 71300 613700 51643.2 \n", + "Activationsd 95900.0 75200 428700 44791.2 \n", + "AsDiscreted 117000.0 90200 4613300 219185.3 \n", + "\n", + " avg% std% min% max% \n", + "Range \n", + "ToTensord 0.222186 0.110528 0.126548 1.332357 \n", + "Lambdad 0.090922 0.042896 0.053021 0.269908 \n", + "GridSplitd 2.066044 1.481407 1.267143 10.711323 \n", + "TorchVisiond_ColorJitter 93.366500 66.815037 34.868515 373.292740 \n", + "ToNumpyd 0.377073 0.161072 0.237763 1.377619 \n", + "RandFlipd 0.162261 0.283646 0.014040 5.077440 \n", + "RandRotate90d 0.242289 0.228988 0.032145 1.940896 \n", + "CastToTyped 0.212154 0.111142 0.124701 2.131919 \n", + "RandZoomd 1.577570 4.087658 0.038796 41.544340 \n", + "ScaleIntensityRanged 1.005373 0.999911 0.594499 16.031329 \n", + "ToTensord_2 0.208038 0.095407 0.131721 1.133760 \n", + "Activationsd 0.201025 0.082748 0.138926 0.791988 \n", + "AsDiscreted 0.268563 0.404927 0.166637 8.522688 " ] }, "execution_count": 7, @@ -434,10 +565,16 @@ "\n", "# Get the entries for training transforms only (to avoid nested ranges)\n", "summary = summary.loc[transforms]\n", - "summary.columns = [c.replace(\" (ns)\", \"\") for c in summary.columns]\n", "\n", - "# Normalize each transform range with total average time (percentage of transfom time)\n", - "summary = summary[[\"Average\", \"Minimum\", \"Maximum\", \"StdDev\"]] / summary[\"Average\"].sum() * 100\n", + "# Nsys output column names are different in different versions, \n", + "# so we need to find the corresponding columns\n", + "avg_col = [c for c in summary.columns if c.startswith('Av')] \n", + "std_col = [c for c in summary.columns if c.startswith('Std')]\n", + "min_col = [c for c in summary.columns if c.startswith('Min')] \n", + "max_col = [c for c in summary.columns if c.startswith('Max')] \n", + "\n", + "# Normalize each transform range with total average time (percentage of transform time)\n", + "summary[[\"avg%\", \"std%\", 'min%', 'max%']] = summary[avg_col+std_col+min_col+max_col] / summary[avg_col].sum()[0] * 100\n", "summary" ] }, @@ -458,7 +595,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -471,8 +608,8 @@ "plt.style.use(\"fivethirtyeight\")\n", "plt.style.use(\"tableau-colorblind10\")\n", "axes = summary.plot.barh(\n", - " y=\"Average\",\n", - " xerr=\"StdDev\",\n", + " y=\"avg%\",\n", + " xerr=\"std%\",\n", " title=\"Average Time\",\n", " xlabel=\"\",\n", " fontsize=16,\n", @@ -502,7 +639,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -513,7 +650,7 @@ ], "source": [ "axes = summary.plot.barh(\n", - " y=[\"Average\", \"Minimum\", \"Maximum\"],\n", + " y=['avg%', \"min%\", \"max%\"],\n", " xlabel=\"\",\n", " fontsize=16,\n", " figsize=(15, 15),\n", @@ -552,7 +689,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.10" + "version": "3.8.12" } }, "nbformat": 4, diff --git a/pathology/tumor_detection/torch/camelyon_train_evaluate_pytorch_gpu.py b/pathology/tumor_detection/torch/camelyon_train_evaluate_pytorch_gpu.py index c6bb4c110e..2fcf6f0cbb 100644 --- a/pathology/tumor_detection/torch/camelyon_train_evaluate_pytorch_gpu.py +++ b/pathology/tumor_detection/torch/camelyon_train_evaluate_pytorch_gpu.py @@ -7,27 +7,28 @@ import monai import numpy as np -from monai.apps.pathology.data import PatchWSIDataset -from monai.data import DataLoader, load_decathlon_datalist +from monai.data import CSVDataset, DataLoader, PatchWSIDataset from monai.networks.nets import TorchVisionFCModel from monai.optimizers import Novograd from monai.transforms import ( Activations, AsDiscrete, CastToType, - CastToTypeD, + CastToTyped, Compose, CuCIM, + GridSplitd, + Lambdad, RandCuCIM, - RandFlipD, - RandRotate90D, - RandZoomD, - ScaleIntensityRangeD, + RandFlipd, + RandRotate90d, + RandZoomd, + ScaleIntensityRanged, ToCupy, - ToNumpyD, - TorchVisionD, + ToNumpyd, + TorchVisiond, ToTensor, - ToTensorD, + ToTensord, ) from monai.utils import first, set_determinism @@ -185,7 +186,7 @@ def main(cfg): fh.setLevel(logging.INFO) logger.addHandler(fh) - # Set TensorBoard summary writter + # Set TensorBoard summary writer writer = SummaryWriter(log_dir) # Save configs @@ -210,11 +211,28 @@ def main(cfg): preprocess_cpu_valid = None preprocess_gpu_valid = None if cfg["backend"] == "cucim": - preprocess_cpu_train = Compose([ToTensorD(keys="label")]) + preprocess_cpu_train = Compose( + [ + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + ToTensord(keys="label"), + ] + ) preprocess_gpu_train = Compose( [ ToCupy(), - RandCuCIM(name="rand_color_jitter", prob=cfg["prob"], brightness=64.0 / 255.0, contrast=0.75, saturation=0.25, hue=0.04), + RandCuCIM( + name="rand_color_jitter", + prob=1.0, + brightness=64.0 / 255.0, + contrast=0.75, + saturation=0.25, + hue=0.04, + ), RandCuCIM(name="rand_image_flip", prob=cfg["prob"], spatial_axis=-1), RandCuCIM(name="rand_image_rotate_90", prob=cfg["prob"], max_k=3, spatial_axis=(-2, -1)), CastToType(dtype=np.float32), @@ -223,7 +241,17 @@ def main(cfg): ToTensor(device=device), ] ) - preprocess_cpu_valid = Compose([ToTensorD(keys="label")]) + preprocess_cpu_valid = Compose( + [ + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + ToTensord(keys="label"), + ] + ) preprocess_gpu_valid = Compose( [ ToCupy(dtype=np.float32), @@ -234,24 +262,36 @@ def main(cfg): elif cfg["backend"] == "numpy": preprocess_cpu_train = Compose( [ - ToTensorD(keys=("image", "label")), - TorchVisionD( + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + ToTensord(keys=("image", "label")), + TorchVisiond( keys="image", name="ColorJitter", brightness=64.0 / 255.0, contrast=0.75, saturation=0.25, hue=0.04 ), - ToNumpyD(keys="image"), - RandFlipD(keys="image", prob=cfg["prob"], spatial_axis=-1), - RandRotate90D(keys="image", prob=cfg["prob"]), - CastToTypeD(keys="image", dtype=np.float32), - RandZoomD(keys="image", prob=cfg["prob"], min_zoom=0.9, max_zoom=1.1), - ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), - ToTensorD(keys="image"), + ToNumpyd(keys="image"), + RandFlipd(keys="image", prob=cfg["prob"], spatial_axis=-1), + RandRotate90d(keys="image", prob=cfg["prob"]), + CastToTyped(keys="image", dtype=np.float32), + RandZoomd(keys="image", prob=cfg["prob"], min_zoom=0.9, max_zoom=1.1), + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensord(keys="image"), ] ) preprocess_cpu_valid = Compose( [ - CastToTypeD(keys="image", dtype=np.float32), - ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), - ToTensorD(keys=("image", "label")), + Lambdad(keys="label", func=lambda x: x.reshape((1, cfg["grid_shape"], cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), + grid=(cfg["grid_shape"], cfg["grid_shape"]), + size={"image": cfg["patch_size"], "label": 1}, + ), + CastToTyped(keys="image", dtype=np.float32), + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensord(keys=("image", "label")), ] ) else: @@ -265,41 +305,42 @@ def main(cfg): ] ) - # Create MONAI dataset - train_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="training", - base_dir=cfg["data_root"], - ) - valid_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="validation", - base_dir=cfg["data_root"], + # Create train dataset and dataloader + train_data_list = CSVDataset( + cfg["train_file"], + col_groups={"image": 0, "patch_location": [2, 1], "label": [3, 6, 9, 4, 7, 10, 5, 8, 11]}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "training/images", x + ".tif")), ) train_dataset = PatchWSIDataset( - data=train_json_info_list, - region_size=cfg["region_size"], - grid_shape=cfg["grid_shape"], - patch_size=cfg["patch_size"], + data=train_data_list, + patch_size=cfg["region_size"], + patch_level=0, transform=preprocess_cpu_train, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", - ) - valid_dataset = PatchWSIDataset( - data=valid_json_info_list, - region_size=cfg["region_size"], - grid_shape=cfg["grid_shape"], - patch_size=cfg["patch_size"], - transform=preprocess_cpu_valid, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", + reader="openslide" if cfg["use_openslide"] else "cuCIM", ) - - # DataLoaders train_dataloader = DataLoader( train_dataset, num_workers=cfg["num_workers"], batch_size=cfg["batch_size"], pin_memory=cfg["pin"] ) - valid_dataloader = DataLoader( - valid_dataset, num_workers=cfg["num_workers"], batch_size=cfg["batch_size"], pin_memory=cfg["pin"] - ) + + # Create validation dataset and dataloader + if not cfg["no_validate"]: + valid_data_list = CSVDataset( + cfg["valid_file"], + col_groups={"image": 0, "patch_location": [2, 1], "label": [3, 6, 9, 4, 7, 10, 5, 8, 11]}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "training/images", x + ".tif")), + ) + valid_dataset = PatchWSIDataset( + data=valid_data_list, + patch_size=cfg["region_size"], + patch_level=0, + transform=preprocess_cpu_valid, + reader="openslide" if cfg["use_openslide"] else "cuCIM", + ) + valid_dataloader = DataLoader( + valid_dataset, num_workers=cfg["num_workers"], batch_size=cfg["batch_size"], pin_memory=cfg["pin"] + ) # Get sample batch and some info first_sample = first(train_dataloader) @@ -314,7 +355,8 @@ def main(cfg): ) logging.info(f"Batch size: {cfg['batch_size']}") logging.info(f"[Training] number of batches: {len(train_dataloader)}") - logging.info(f"[Validation] number of batches: {len(valid_dataloader)}") + if not cfg["no_validate"]: + logging.info(f"[Validation] number of batches: {len(valid_dataloader)}") # ------------------------------------------------------------------------- # Deep Learning Model and Configurations # ------------------------------------------------------------------------- @@ -352,7 +394,10 @@ def main(cfg): total_valid_time, total_train_time = 0.0, 0.0 t_start = time.perf_counter() - metric_summary = {"loss": np.Inf, "accuracy": 0, "best_epoch": 1} + if cfg["no_validate"]: + metric_summary = {} + else: + metric_summary = {"loss": np.Inf, "accuracy": 0, "best_epoch": 1} # Training/Validation Loop for _ in range(cfg["n_epochs"]): t_epoch = time.perf_counter() @@ -375,14 +420,16 @@ def main(cfg): ) if scheduler is not None: scheduler.step() - if cfg["save"]: + if not cfg["no_save"]: torch.save(model.state_dict(), os.path.join(log_dir, f"model_epoch_{train_counter['epoch']}.pt")) t_train = time.perf_counter() train_time = t_train - t_epoch total_train_time += train_time # Validation - if cfg["validate"]: + if cfg["no_validate"]: + logging.info(f"[Epoch: {train_counter['epoch']}/{cfg['n_epochs']}] Train time: {train_time:.1f}s") + else: valid_loss, valid_acc = validation( model, loss_func, @@ -407,8 +454,6 @@ def main(cfg): f"[Epoch: {train_counter['epoch']}/{cfg['n_epochs']}] loss: {valid_loss:.3f}, accuracy: {valid_acc:.3f}, " f"time: {t_valid - t_epoch:.1f}s (train: {train_time:.1f}s, valid: {valid_time:.1f}s)" ) - else: - logging.info(f"[Epoch: {train_counter['epoch']}/{cfg['n_epochs']}] Train time: {train_time:.1f}s") writer.flush() t_end = time.perf_counter() @@ -420,7 +465,7 @@ def main(cfg): logging.info(f"Metric Summary: {metric_summary}") # Save the best and final model - if cfg["validate"] is True: + if not cfg["no_validate"] and not cfg["no_save"]: copyfile( os.path.join(log_dir, f"model_epoch_{metric_summary['best_epoch']}.pt"), os.path.join(log_dir, "model_best.pt"), @@ -440,24 +485,18 @@ def main(cfg): def parse_arguments(): parser = ArgumentParser(description="Tumor detection on whole slide pathology images.") - parser.add_argument( - "--dataset", - type=str, - default="./dataset_0.json", - dest="dataset_json", - help="path to dataset json file", - ) parser.add_argument( "--root", type=str, - default="/workspace/data/medical/pathology/", - dest="data_root", - help="path to root folder of images containing training folder", + default="/workspace/data/medical/pathology", + help="path to image folder containing training/validation", ) + parser.add_argument("--train-file", type=str, default="training.csv", help="path to training data file") + parser.add_argument("--valid-file", type=str, default="validation.csv", help="path to training data file") parser.add_argument("--logdir", type=str, default="./logs/", dest="logdir", help="log directory") parser.add_argument("--rs", type=int, default=256 * 3, dest="region_size", help="region size") - parser.add_argument("--gs", type=int, default=3, dest="grid_shape", help="image grid shape (3x3)") + parser.add_argument("--gs", type=int, default=3, dest="grid_shape", help="image grid shape e.g 3 means 3x3") parser.add_argument("--ps", type=int, default=224, dest="patch_size", help="patch size") parser.add_argument("--bs", type=int, default=64, dest="batch_size", help="batch size") parser.add_argument("--ep", type=int, default=4, dest="n_epochs", help="number of epochs") @@ -473,8 +512,8 @@ def parse_arguments(): parser.add_argument("--pretrain", action="store_true", help="activate Imagenet weights") parser.add_argument("--benchmark", action="store_true", help="activate Imagenet weights") - parser.add_argument("--save", action="store_true", help="save model at each epoch") - parser.add_argument("--validate", action="store_true", help="use optimized parameters") + parser.add_argument("--no-save", action="store_true", help="save model at each epoch") + parser.add_argument("--no-validate", action="store_true", help="use optimized parameters") parser.add_argument("--baseline", action="store_true", help="use baseline parameters") parser.add_argument("--optimized", action="store_true", help="use optimized parameters") parser.add_argument("-b", "--backend", type=str, dest="backend", help="backend for transforms") diff --git a/performance_profiling/pathology/train_evaluate_nvtx.py b/performance_profiling/pathology/train_evaluate_nvtx.py index dbe3d77130..344681a245 100644 --- a/performance_profiling/pathology/train_evaluate_nvtx.py +++ b/performance_profiling/pathology/train_evaluate_nvtx.py @@ -7,27 +7,28 @@ import monai import numpy as np -from monai.apps.pathology.data import PatchWSIDataset -from monai.data import DataLoader, load_decathlon_datalist +from monai.data import CSVDataset, DataLoader, PatchWSIDataset from monai.networks.nets import TorchVisionFCModel from monai.optimizers import Novograd from monai.transforms import ( Activations, AsDiscrete, CastToType, - CastToTypeD, + CastToTyped, Compose, CuCIM, + GridSplitd, + Lambdad, RandCuCIM, - RandFlipD, - RandRotate90D, - RandZoomD, - ScaleIntensityRangeD, + RandFlipd, + RandRotate90d, + RandZoomd, + ScaleIntensityRanged, ToCupy, - ToNumpyD, - TorchVisionD, + ToNumpyd, + TorchVisiond, ToTensor, - ToTensorD, + ToTensord, ) from monai.utils import first, set_determinism, Range @@ -78,20 +79,23 @@ def training( writer: SummaryWriter, print_step, ): + summary["epoch"] += 1 + model.train() n_steps = len(dataloader) iter_data = iter(dataloader) for step in range(n_steps): + summary["step"] += 1 with Range("Step"): with Range("Data Loading"): batch = next(iter_data) x = batch["image"].to(device) y = batch["label"].to(device) - if pre_process is not None: - x = pre_process(x) + if pre_process is not None: + x = pre_process(x) with autocast(enabled=amp): output = model(x) @@ -121,10 +125,6 @@ def training( f"Step: {step + 1}/{n_steps} -- " f"train_loss: {loss_data:.5f}, train_acc: {acc_data:.3f}" ) - - summary["step"] += 1 - - summary["epoch"] += 1 return summary @@ -212,41 +212,66 @@ def main(cfg): preprocess_cpu_valid = None preprocess_gpu_valid = None if cfg["backend"] == "cucim": - preprocess_cpu_train = Compose([ToTensorD(keys="label")]) + preprocess_cpu_train = Compose( + [ + Lambdad(keys="label", func=lambda x: x.reshape((1, *cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), grid=cfg["grid_shape"], size={"image": cfg["patch_size"], "label": 1} + ), + ToTensord(keys="label"), + ] + ) preprocess_gpu_train = Compose( [ Range()(ToCupy()), Range("ColorJitter")( - RandCuCIM(name="color_jitter", brightness=64.0 / 255.0, contrast=0.75, saturation=0.25, hue=0.04) + RandCuCIM( + name="rand_color_jitter", + prob=1.0, + brightness=64.0 / 255.0, + contrast=0.75, + saturation=0.25, + hue=0.04, + ) ), - Range("RandomFlip")(RandCuCIM(name="image_flip", apply_prob=cfg["prob"], spatial_axis=-1)), + Range("RandomFlip")(RandCuCIM(name="rand_image_flip", prob=cfg["prob"], spatial_axis=-1)), Range("RandomRotate90")( RandCuCIM(name="rand_image_rotate_90", prob=cfg["prob"], max_k=3, spatial_axis=(-2, -1)) ), Range()(CastToType(dtype=np.float32)), - Range("RandomZoom")(RandCuCIM(name="rand_zoom", min_zoom=0.9, max_zoom=1.1)), + Range("RandomZoom")(RandCuCIM(name="rand_zoom", prob=cfg["prob"], min_zoom=0.9, max_zoom=1.1)), Range("ScaleIntensity")( CuCIM(name="scale_intensity_range", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0) ), Range()(ToTensor(device=device)), ] ) - preprocess_cpu_valid = Compose([ToTensorD(keys="label")]) - preprocess_gpu_valid = Compose( + preprocess_cpu_valid = Compose( [ - Range("ValidToCupyAndCast")(ToCupy(dtype=np.float32)), - Range("ValidScaleIntensity")( - CuCIM(name="scale_intensity_range", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0) + Lambdad(keys="label", func=lambda x: x.reshape((1, *cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), grid=cfg["grid_shape"], size={"image": cfg["patch_size"], "label": 1} ), - Range("ValidToTensor")(ToTensor(device=device)), + ToTensord(keys="label"), + ] + ) + preprocess_gpu_valid = Compose( + [ + ToCupy(dtype=np.float32), + CuCIM(name="scale_intensity_range", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensor(device=device), ] ) elif cfg["backend"] == "numpy": preprocess_cpu_train = Compose( [ - Range()(ToTensorD(keys=("image", "label"))), + Lambdad(keys="label", func=lambda x: x.reshape((1, *cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), grid=cfg["grid_shape"], size={"image": cfg["patch_size"], "label": 1} + ), + Range()(ToTensord(keys=("image", "label"))), Range("ColorJitter")( - TorchVisionD( + TorchVisiond( keys="image", name="ColorJitter", brightness=64.0 / 255.0, @@ -255,24 +280,26 @@ def main(cfg): hue=0.04, ) ), - Range()(ToNumpyD(keys="image")), - Range("RandomFlip")(RandFlipD(keys="image", prob=cfg["prob"], spatial_axis=-1)), - Range("RandomRotate90")(RandRotate90D(keys="image", prob=cfg["prob"])), - Range()(CastToTypeD(keys="image", dtype=np.float32)), - Range("RandomZoom")(RandZoomD(keys="image", prob=cfg["prob"], min_zoom=0.9, max_zoom=1.1)), + Range()(ToNumpyd(keys="image")), + Range("RandomFlip")(RandFlipd(keys="image", prob=cfg["prob"], spatial_axis=-1)), + Range("RandomRotate90")(RandRotate90d(keys="image", prob=cfg["prob"])), + Range()(CastToTyped(keys="image", dtype=np.float32)), + Range("RandomZoom")(RandZoomd(keys="image", prob=cfg["prob"], min_zoom=0.9, max_zoom=1.1)), Range("ScaleIntensity")( - ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0) + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0) ), - Range()(ToTensorD(keys="image")), + Range()(ToTensord(keys="image")), ] ) preprocess_cpu_valid = Compose( [ - Range("ValidCastType")(CastToTypeD(keys="image", dtype=np.float32)), - Range("ValidScaleIntensity")( - ScaleIntensityRangeD(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0) + Lambdad(keys="label", func=lambda x: x.reshape((1, *cfg["grid_shape"]))), + GridSplitd( + keys=("image", "label"), grid=cfg["grid_shape"], size={"image": cfg["patch_size"], "label": 1} ), - Range("ValidToTensor")(ToTensorD(keys=("image", "label"))), + CastToTyped(keys="image", dtype=np.float32), + ScaleIntensityRanged(keys="image", a_min=0.0, a_max=255.0, b_min=-1.0, b_max=1.0), + ToTensord(keys=("image", "label")), ] ) else: @@ -287,31 +314,32 @@ def main(cfg): ) # Create MONAI dataset - train_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="training", - base_dir=cfg["data_root"], - ) - valid_json_info_list = load_decathlon_datalist( - data_list_file_path=cfg["dataset_json"], - data_list_key="validation", - base_dir=cfg["data_root"], + train_data_list = CSVDataset( + cfg["train_file"], + col_groups={"image": 0, "location": [2, 1], "label": list(range(3, 12))}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "training/images", x + ".tif")), ) train_dataset = PatchWSIDataset( - data=train_json_info_list, - region_size=cfg["region_size"], - grid_shape=cfg["grid_shape"], - patch_size=cfg["patch_size"], + data=train_data_list, + size=cfg["region_size"], + level=0, transform=preprocess_cpu_train, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", + reader="openslide" if cfg["use_openslide"] else "cuCIM", + ) + + valid_data_list = CSVDataset( + cfg["valid_file"], + col_groups={"image": 0, "location": [2, 1], "label": list(range(3, 12))}, + kwargs_read_csv={"header": None}, + transform=Lambdad("image", lambda x: os.path.join(cfg["root"], "validation/images", x + ".tif")), ) valid_dataset = PatchWSIDataset( - data=valid_json_info_list, - region_size=cfg["region_size"], - grid_shape=cfg["grid_shape"], - patch_size=cfg["patch_size"], + data=valid_data_list, + size=cfg["region_size"], + level=0, transform=preprocess_cpu_valid, - image_reader_name="openslide" if cfg["use_openslide"] else "cuCIM", + reader="openslide" if cfg["use_openslide"] else "cuCIM", ) # DataLoaders @@ -369,7 +397,7 @@ def main(cfg): # ------------------------------------------------------------------------- # Training/Evaluating # ------------------------------------------------------------------------- - train_counter = {"n_epochs": cfg["n_epochs"], "epoch": 1, "step": 1} + train_counter = {"n_epochs": cfg["n_epochs"], "epoch": 0, "step": 0} total_valid_time, total_train_time = 0.0, 0.0 t_start = time.perf_counter() @@ -427,7 +455,7 @@ def main(cfg): writer.add_scalar("valid/accuracy", valid_acc, train_counter["epoch"]) logging.info( - f"[Epoch: {train_counter['epoch']}/{cfg['n_epochs']}] loss: {valid_loss:.3f}, accuracy: {valid_acc:.2f}, " + f"[Epoch: {train_counter['epoch']}/{cfg['n_epochs']}] loss: {valid_loss:.3f}, accuracy: {valid_acc:.3f}, " f"time: {t_valid - t_epoch:.1f}s (train: {train_time:.1f}s, valid: {valid_time:.1f}s)" ) else: @@ -445,12 +473,12 @@ def main(cfg): # Save the best and final model if cfg["validate"] is True: copyfile( - os.path.join(log_dir, f"model_epoch_{metric_summary['best_epoch']}.pth"), - os.path.join(log_dir, "model_best.pth"), + os.path.join(log_dir, f"model_epoch_{metric_summary['best_epoch']}.pt"), + os.path.join(log_dir, "model_best.pt"), ) copyfile( - os.path.join(log_dir, f"model_epoch_{cfg['n_epochs']}.pth"), - os.path.join(log_dir, "model_final.pth"), + os.path.join(log_dir, f"model_epoch_{cfg['n_epochs']}.pt"), + os.path.join(log_dir, "model_final.pt"), ) # Final prints @@ -463,24 +491,13 @@ def main(cfg): def parse_arguments(): parser = ArgumentParser(description="Tumor detection on whole slide pathology images.") - parser.add_argument( - "--dataset", - type=str, - default="./data/dataset_0.json", - dest="dataset_json", - help="path to dataset json file", - ) - parser.add_argument( - "--root", - type=str, - default="data/", - dest="data_root", - help="path to root folder of images containing training folder", - ) + parser.add_argument("--root", type=str, default="./", help="path to image folder containing training/validation") + parser.add_argument("--train-file", type=str, default="training.csv", help="path to training data file") + parser.add_argument("--valid-file", type=str, default="validation.csv", help="path to training data file") parser.add_argument("--logdir", type=str, default="./logs/", dest="logdir", help="log directory") parser.add_argument("--rs", type=int, default=256 * 3, dest="region_size", help="region size") - parser.add_argument("--gs", type=int, default=3, dest="grid_shape", help="image grid shape (3x3)") + parser.add_argument("--gs", type=int, default=(3, 3), nargs="+", dest="grid_shape", help="image grid shape (3x3)") parser.add_argument("--ps", type=int, default=224, dest="patch_size", help="patch size") parser.add_argument("--bs", type=int, default=64, dest="batch_size", help="batch size") parser.add_argument("--ep", type=int, default=4, dest="n_epochs", help="number of epochs") @@ -502,7 +519,7 @@ def parse_arguments(): parser.add_argument("--optimized", action="store_true", help="use optimized parameters") parser.add_argument("-b", "--backend", type=str, dest="backend", help="backend for transforms") - parser.add_argument("--cpu", type=int, default=10, dest="num_workers", help="number of workers") + parser.add_argument("--cpu", type=int, default=8, dest="num_workers", help="number of workers") parser.add_argument("--gpu", type=str, default="0", dest="gpu", help="which gpu to use") args = parser.parse_args() @@ -521,12 +538,12 @@ def parse_arguments(): config_dict["backend"] = "cucim" if config_dict["baseline"] is True: - config_dict["benchmark"] = True - config_dict["novograd"] = True + config_dict["benchmark"] = False + config_dict["novograd"] = False config_dict["pretrain"] = True - config_dict["cos"] = True + config_dict["cos"] = False config_dict["pin"] = False - config_dict["amp"] = True + config_dict["amp"] = False if config_dict["backend"] is None: config_dict["backend"] = "numpy"