[pre-commit.ci] pre-commit suggestions (Project-MONAI#1280)

<!--pre-commit.ci start-->
updates:
- [github.com/psf/black: 22.12.0 →
23.3.0](psf/black@22.12.0...23.3.0)
<!--pre-commit.ci end-->

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Wenqi Li <[email protected]>

.pre-commit-config.yaml

@@ -23,7 +23,7 @@ repos:
       - id: detect-private-key
 
   - repo: https://github.com/psf/black
-    rev: "22.12.0"
+    rev: "23.3.0"
     hooks:
       - id: black
       - id: black-jupyter

3d_classification/ignite/densenet_training_dict.py

@@ -93,7 +93,6 @@ def main():
     # Ignite trainer expects batch=(img, label) and returns output=loss at every iteration,
     # user can add output_transform to return other values, like: y_pred, y, etc.
     def prepare_batch(batch, device=None, non_blocking=False):
-
         return _prepare_batch((batch["img"], batch["label"]), device, non_blocking)
 
     trainer = create_supervised_trainer(net, opt, loss, device, False, prepare_batch=prepare_batch)

3d_registration/paired_lung_ct.ipynb

@@ -676,7 +676,6 @@
     "        model.eval()\n",
     "        with torch.no_grad():\n",
     "            for val_data in val_loader:\n",
-    "\n",
     "                val_ddf, val_pred_image, val_pred_label = forward(val_data, model)\n",
     "                val_pred_label[val_pred_label > 1] = 1\n",
     "\n",

3d_segmentation/brats_segmentation_3d.ipynb

@@ -541,7 +541,6 @@
     "    if (epoch + 1) % val_interval == 0:\n",
     "        model.eval()\n",
     "        with torch.no_grad():\n",
-    "\n",
     "            for val_data in val_loader:\n",
     "                val_inputs, val_labels = (\n",
     "                    val_data[\"image\"].to(device),\n",

3d_segmentation/swin_unetr_brats21_segmentation_3d.ipynb

@@ -268,7 +268,6 @@
     "\n",
     "\n",
     "def datafold_read(datalist, basedir, fold=0, key=\"training\"):\n",
-    "\n",
     "    with open(datalist) as f:\n",
     "        json_data = json.load(f)\n",
     "\n",
@@ -610,7 +609,6 @@
     "    post_sigmoid=None,\n",
     "    post_pred=None,\n",
     "):\n",
-    "\n",
     "    val_acc_max = 0.0\n",
     "    dices_tc = []\n",
     "    dices_wt = []\n",
@@ -705,7 +703,15 @@
    "source": [
     "start_epoch = 0\n",
     "\n",
-    "(val_acc_max, dices_tc, dices_wt, dices_et, dices_avg, loss_epochs, trains_epoch,) = trainer(\n",
+    "(\n",
+    "    val_acc_max,\n",
+    "    dices_tc,\n",
+    "    dices_wt,\n",
+    "    dices_et,\n",
+    "    dices_avg,\n",
+    "    loss_epochs,\n",
+    "    trains_epoch,\n",
+    ") = trainer(\n",
     "    model=model,\n",
     "    train_loader=train_loader,\n",
     "    val_loader=val_loader,\n",

acceleration/distributed_training/brats_training_ddp.py

@@ -132,7 +132,6 @@ def __init__(
         num_workers=0,
         shuffle=False,
     ) -> None:
-
         if not os.path.isdir(root_dir):
             raise ValueError("root directory root_dir must be a directory.")
         self.section = section

auto3dseg/notebooks/ensemble_byoc.ipynb

@@ -175,7 +175,6 @@
     "    \"\"\"\n",
     "\n",
     "    def __init__(self, n_models=3):\n",
-    "\n",
     "        super().__init__()\n",
     "        self.n_models = n_models\n",
     "\n",

auto3dseg/tasks/hecktor22/hecktor_crop_neck_region.py

@@ -37,7 +37,6 @@ def __init__(
         self.box_size = box_size
 
     def __call__(self, data):
-
         d = dict(data)
 
         im_pet = d["image2"][0]
@@ -92,7 +91,6 @@ def __call__(self, data):
         return d
 
     def extract_roi(self, im_pet, box_size):
-
         crop_len = int(0.75 * im_pet.shape[2])
         im = im_pet[..., crop_len:]
 

competitions/MICCAI/surgtoolloc/classification_files/train.py

@@ -35,7 +35,6 @@
 
 
 def main(cfg):
-
     os.makedirs(str(cfg.output_dir + f"/fold{cfg.fold}/"), exist_ok=True)
     set_seed(cfg.seed)
     # set dataset, dataloader
@@ -190,7 +189,6 @@ def run_train(
 
 
 def run_eval(model, val_dataloader, cfg, writer, epoch, metric):
-
     model.eval()
     torch.set_grad_enabled(False)
 
@@ -213,7 +211,6 @@ def run_eval(model, val_dataloader, cfg, writer, epoch, metric):
 
 
 if __name__ == "__main__":
-
     sys.path.append("configs")
     sys.path.append("models")
     sys.path.append("data")

competitions/MICCAI/surgtoolloc/classification_files/utils.py

@@ -45,7 +45,6 @@ def set_seed(seed):
 
 
 def get_train_dataloader(train_dataset, cfg):
-
     train_dataloader = DataLoader(
         train_dataset,
         sampler=None,
@@ -61,7 +60,6 @@ def get_train_dataloader(train_dataset, cfg):
 
 
 def get_val_dataloader(val_dataset, cfg):
-
     val_dataloader = DataLoader(
         val_dataset,
         shuffle=False,

competitions/kaggle/RANZCR/4th_place_solution/data/seg_data.py

@@ -97,7 +97,6 @@ def get_mask(self, study_id: str, img_shape: Tuple, is_annotated: int):
         mask = np.zeros((img_shape[0], img_shape[1], self.cfg.seg_dim))
 
         for idx, data in df.iterrows():
-
             xys = [np.array(ast.literal_eval(data["data"])).clip(0, np.inf).astype(np.int32)[:, None, :]]
 
             m = np.zeros(img_shape)
@@ -120,7 +119,6 @@ def get_mask(self, study_id: str, img_shape: Tuple, is_annotated: int):
         return mask
 
     def __getitem__(self, idx):
-
         study_id = self.study_ids[idx]
         label = self.labels[idx]
         is_annotated = self.is_annotated[idx]

competitions/kaggle/RANZCR/4th_place_solution/models/seg_model.py

@@ -92,7 +92,6 @@ def __init__(
         self.blocks = nn.ModuleList(blocks)
 
     def forward(self, *features: Sequence[torch.Tensor]):
-
         features = features[1:][::-1]
         skips = features[1:]
         x = features[0]
@@ -215,7 +214,6 @@ def __init__(self, cfg):
             print("weights loaded from", cfg.pretrained_weights)
 
     def forward(self, batch):
-
         x_in = batch["input"]
         enc_out = self.encoder(x_in)
 

competitions/kaggle/RANZCR/4th_place_solution/train.py

@@ -42,7 +42,6 @@
 
 
 def main(cfg):
-
     os.makedirs(str(cfg.output_dir + f"/fold{cfg.fold}/"), exist_ok=True)
 
     # set random seed, works when use all data to train
@@ -194,12 +193,10 @@ def run_train(
             scheduler.step()
 
         if step % cfg.batch_size == 0:
-
             progress_bar.set_description(f"loss: {np.mean(losses[-10:]):.2f}")
 
 
 def run_eval(model, val_dataloader, cfg, writer, epoch):
-
     model.eval()
     torch.set_grad_enabled(False)
 
@@ -228,7 +225,6 @@ def run_eval(model, val_dataloader, cfg, writer, epoch):
     val_loss = np.mean(val_losses)
 
     if cfg.compute_auc is True:
-
         val_preds = torch.cat(val_preds)
         val_targets = torch.cat(val_targets)
         avg_auc = compute_roc_auc(val_preds, val_targets, average="macro")
@@ -240,7 +236,6 @@ def run_eval(model, val_dataloader, cfg, writer, epoch):
 
 
 def run_infer(weights_folder_path, cfg):
-
     cfg.pretrained = False
     # for local test, please modify the following path into actual path.
     cfg.data_folder = cfg.data_dir + "test/"
@@ -296,7 +291,6 @@ def run_infer(weights_folder_path, cfg):
 
 
 if __name__ == "__main__":
-
     sys.path.append("configs")
     sys.path.append("models")
     sys.path.append("data")

competitions/kaggle/RANZCR/4th_place_solution/utils.py

@@ -33,7 +33,6 @@ def get_train_dataset(train_df, cfg):
 
 
 def get_train_dataloader(train_dataset, cfg):
-
     train_dataloader = DataLoader(
         train_dataset,
         sampler=None,
@@ -54,7 +53,6 @@ def get_val_dataset(val_df, cfg):
 
 
 def get_val_dataloader(val_dataset, cfg):
-
     val_dataloader = DataLoader(
         val_dataset,
         sampler=SequentialSampler(val_dataset),
@@ -73,7 +71,6 @@ def get_test_dataset(test_df, cfg):
 
 
 def get_test_dataloader(test_dataset, cfg):
-
     test_dataloader = DataLoader(
         test_dataset,
         shuffle=False,
@@ -84,15 +81,13 @@ def get_test_dataloader(test_dataset, cfg):
 
 
 def get_optimizer(model, cfg):
-
     params = model.parameters()
     optimizer = optim.Adam(params, lr=cfg.lr, weight_decay=cfg.weight_decay)
 
     return optimizer
 
 
 def get_scheduler(cfg, optimizer, total_steps):
-
     scheduler = WarmupCosineSchedule(
         optimizer,
         warmup_steps=cfg.warmup * (total_steps // cfg.batch_size),

deep_atlas/deep_atlas_tutorial.ipynb

@@ -1251,7 +1251,6 @@
     "val_interval = 5\n",
     "\n",
     "for epoch_number in range(max_epochs):\n",
-    "\n",
     "    print(f\"Epoch {epoch_number+1}/{max_epochs}:\")\n",
     "\n",
     "    seg_net.train()\n",
@@ -1794,7 +1793,6 @@
     "best_reg_validation_loss = float(\"inf\")\n",
     "\n",
     "for epoch_number in range(max_epochs):\n",
-    "\n",
     "    print(f\"Epoch {epoch_number+1}/{max_epochs}:\")\n",
     "\n",
     "    # ------------------------------------------------\n",

deepedit/ignite/train.py

@@ -194,7 +194,6 @@ def get_loaders(args, pre_transforms):
 
 
 def create_trainer(args):
-
     set_determinism(seed=args.seed)
 
     multi_gpu = args.multi_gpu
@@ -431,7 +430,6 @@ def main():
 
 
 if __name__ == "__main__":
-
     logging.basicConfig(
         stream=sys.stdout,
         level=logging.INFO,

deployment/Triton/client/client.py

@@ -61,7 +61,6 @@ def open_nifti_files(input_path):
 
 
 if __name__ == "__main__":
-
     parser = argparse.ArgumentParser(description="Triton CLI for COVID classification inference from NIFTI data")
     parser.add_argument(
         "input",

deployment/Triton/client/client_mednist.py

@@ -65,7 +65,6 @@ def open_jpeg_files(input_path):
 
 
 if __name__ == "__main__":
-
     parser = argparse.ArgumentParser(description="Triton CLI for MedNist classification inference from JPEG data")
     parser.add_argument(
         "input",

deployment/Triton/models/mednist_class/1/model.py

@@ -158,7 +158,6 @@ def execute(self, requests):
         batched_img = []
         print("starting request")
         for request in requests:
-
             # get the input by name (as configured in config.pbtxt)
             input_0 = pb_utils.get_input_tensor_by_name(request, "INPUT0")
 

deployment/Triton/models/monai_covid/1/model.py

@@ -152,7 +152,6 @@ def execute(self, requests):
         responses = []
 
         for request in requests:
-
             # get the input by name (as configured in config.pbtxt)
             input_0 = pb_utils.get_input_tensor_by_name(request, "INPUT0")
 

detection/luna16_visualization/save_obj.py

@@ -23,9 +23,7 @@
 
 
 def save_obj(vertices, faces, filename):
-
     with open(filename, "w") as f:
-
         for v in vertices:
             f.write("v {} {} {}\n".format(*np.array(v)))
 

federated_learning/openfl/openfl_mednist_2d_registration/workspace/Monai_MedNIST.ipynb

@@ -380,7 +380,6 @@
     "    loss_fn=image_loss,\n",
     "    affine_transform=warp_layer,\n",
     "):\n",
-    "\n",
     "    train_loader = tqdm.tqdm(train_loader, desc=\"train\")\n",
     "    net_model.train()\n",
     "    net_model.to(device)\n",

federated_learning/substra/assets/objective/metrics.py

@@ -22,7 +22,7 @@ def score(self, y_true, y_pred):
         metric_sum = 0.0
         metric_count = 0
         with torch.no_grad():
-            for (val_true, val_pred) in zip(y_true, y_pred):
+            for val_true, val_pred in zip(y_true, y_pred):
                 val_true, _ = val_true
                 val_pred, _ = val_pred
                 value = self.dice_metric(

model_zoo/TCIA_PROSTATEx_Prostate_MRI_Anatomy_Model.ipynb

@@ -324,6 +324,7 @@
     "# scripts_dir = os.path.join(zoo_dir, model_name, \"scripts\")\n",
     "# sys.path.insert(1, scripts_dir)\n",
     "\n",
+    "\n",
     "# Compact alternative implementation of this model's specific cropping step.\n",
     "#   Ideally this would have been accomplished using MONAI's transforms\n",
     "#   for data pre-processing / augmentation instead of using a separate\n",

modules/autoencoder_mednist.ipynb

@@ -364,7 +364,6 @@
    "outputs": [],
    "source": [
     "def train(dict_key_for_training, max_epochs=10, learning_rate=1e-3):\n",
-    "\n",
     "    model = AutoEncoder(\n",
     "        spatial_dims=2,\n",
     "        in_channels=1,\n",

modules/generate_random_permutations/randomizedPermutations.py

@@ -118,7 +118,6 @@ def __call__(self, image_file_list, *args, **kwargs):
 
 
 def main():
-
     image_dir = "./exampleImages"
     image_file_list = glob.glob(image_dir + "/*.png")
     output_size = (400, 400)

modules/interpretability/class_lung_lesion.ipynb

@@ -605,7 +605,6 @@
     "\n",
     "example = 0\n",
     "for item in items:\n",
-    "\n",
     "    data = train_ds[item]  # this fetches training data with random augmentations\n",
     "    image, label = data[\"image\"].to(device).unsqueeze(0), data[\"label\"][1]\n",
     "    y_pred = model_3d(image)\n",

modules/inverse_transforms_and_test_time_augmentations.ipynb

@@ -729,7 +729,6 @@
     "\n",
     "# Get images\n",
     "for file in np.random.choice(val_files, size=5, replace=False):\n",
-    "\n",
     "    mode_tta, mean_tta, std_tta, vvc_tta = tt_aug(file, num_examples=10)\n",
     "    unmodified_data = minimal_transforms(file)\n",
     "\n",

modules/lazy_resampling_functional.ipynb

@@ -91,7 +91,6 @@
    "outputs": [],
    "source": [
     "def plot_img_operations(img):\n",
-    "\n",
     "    if isinstance(img, list):\n",
     "        print(f\"patch metadata of patch idx 0 out of {len(img)} samples.\\n\")\n",
     "        img_0 = img[0]\n",