[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

Author: pre-commit-ci[bot]

acceleration/fast_inference_tutorial/fast_inference_tutorial.ipynb

@@ -354,7 +354,7 @@
     "    total_time_dict = {}\n",
     "    roi_size = (96, 96, 96)\n",
     "    sw_batch_size = 1\n",
-    "    \n",
+    "\n",
     "    for idx, sample in enumerate(data_list[:10]):\n",
     "        start = timer()\n",
     "        data = infer_transforms({\"image\": sample})\n",
@@ -465,7 +465,9 @@
     "    all_df = pd.merge(all_df, df, on=\"file_name\", how=\"left\")\n",
     "\n",
     "# for each file, add it's size\n",
-    "all_df[\"file_size\"] = all_df[\"file_name\"].apply(lambda x: os.path.getsize(os.path.join(root_dir, \"Task03_Liver\", \"imagesTs_nii\", x)))\n",
+    "all_df[\"file_size\"] = all_df[\"file_name\"].apply(\n",
+    "    lambda x: os.path.getsize(os.path.join(root_dir, \"Task03_Liver\", \"imagesTs_nii\", x))\n",
+    ")\n",
     "# sort by file size\n",
     "all_df = all_df.sort_values(by=\"file_size\", ascending=True)\n",
     "# convert file size to MB\n",
@@ -489,10 +491,10 @@
    "outputs": [],
    "source": [
     "plt.figure(figsize=(10, 6))\n",
-    "average_time.plot(kind='bar', color=['skyblue', 'orange', 'green', 'red'])\n",
-    "plt.title('Average Inference Time for Each Benchmark Type')\n",
-    "plt.xlabel('Benchmark Type')\n",
-    "plt.ylabel('Average Time (seconds)')\n",
+    "average_time.plot(kind=\"bar\", color=[\"skyblue\", \"orange\", \"green\", \"red\"])\n",
+    "plt.title(\"Average Inference Time for Each Benchmark Type\")\n",
+    "plt.xlabel(\"Benchmark Type\")\n",
+    "plt.ylabel(\"Average Time (seconds)\")\n",
     "plt.xticks(rotation=45)\n",
     "plt.tight_layout()\n",
     "plt.show()"

acceleration/fast_inference_tutorial/run_benchmark.py

@@ -20,13 +20,21 @@
 import torch_tensorrt
 from monai.inferers import sliding_window_inference
 from monai.networks.nets import SegResNet
-from monai.transforms import (Activationsd, AsDiscreted, Compose,
-                              EnsureChannelFirstd, EnsureTyped, Invertd,
-                              LoadImaged, NormalizeIntensityd, Orientationd,
-                              ScaleIntensityd, Spacingd)
-
-from utils import (prepare_model_weights, prepare_tensorrt_model,
-                   prepare_test_datalist)
+from monai.transforms import (
+    Activationsd,
+    AsDiscreted,
+    Compose,
+    EnsureChannelFirstd,
+    EnsureTyped,
+    Invertd,
+    LoadImaged,
+    NormalizeIntensityd,
+    Orientationd,
+    ScaleIntensityd,
+    Spacingd,
+)
+
+from utils import prepare_model_weights, prepare_tensorrt_model, prepare_test_datalist
 
 
 def get_transforms(device, gpu_loading_flag=False, gpu_transforms_flag=False):
@@ -49,6 +57,7 @@ def get_transforms(device, gpu_loading_flag=False, gpu_transforms_flag=False):
 
     return infer_transforms
 
+
 def get_post_transforms(infer_transforms):
     post_transforms = Compose(
         [
@@ -65,6 +74,7 @@ def get_post_transforms(infer_transforms):
     )
     return post_transforms
 
+
 def get_model(device, weights_path, trt_model_path, trt_flag=False):
     if not trt_flag:
         model = SegResNet(
@@ -84,11 +94,12 @@ def get_model(device, weights_path, trt_model_path, trt_flag=False):
         model = torch.jit.load(trt_model_path)
     return model
 
+
 def run_inference(data_list, infer_transforms, model, device, benchmark_type):
     total_time_dict = {}
     roi_size = (96, 96, 96)
     sw_batch_size = 1
-    
+
     for idx, sample in enumerate(data_list):
         start = timer()
         data = infer_transforms({"image": sample})
@@ -117,6 +128,7 @@ def run_inference(data_list, infer_transforms, model, device, benchmark_type):
 
     return total_time_dict
 
+
 def main():
     parser = argparse.ArgumentParser(description="Run inference benchmark.")
     parser.add_argument("--benchmark_type", type=str, default="original", help="Type of benchmark to run")
@@ -146,5 +158,6 @@ def main():
     df = pd.DataFrame(list(total_time_dict.items()), columns=["file_name", "time"])
     df.to_csv(os.path.join(root_dir, f"time_{args.benchmark_type}.csv"), index=False)
 
+
 if __name__ == "__main__":
     main()