pytorch/vision

15.4

Adjusted Score

15.4

Raw Score

100%

Time Factor

2026-07-14

Last Push

17.8K

Stars

Python

Language

119.5K

Lines of Code

431

Files

1.7K

Pattern Hits

2026-07-14

Scan Date

0.18

HC Hit Rate

What These Metrics Mean

Adjusted Score: Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score: The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor: The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits: Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate: High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files: Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.

Severity Breakdown

Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.

CRITICAL 0HIGH 79MEDIUM 66LOW 1513

Directory Score Breakdown

This horizontal bar chart decomposes the repository's raw synthetic code score by top-level directory, allowing you to pinpoint precisely which modules or components carry the highest AI authorship density. Directories with disproportionately high scores relative to their size warrant targeted manual review: concentrated AI signatures often trace back to mass-generated configuration layers, auto-ported test suites, LLM-scaffolded boilerplate classes, or entire subsystems authored under heavy copilot assistance. Use this view to prioritise your human code-review effort.

Pattern Findings

The scanner identified 1658 distinct pattern matches across 17 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the diagnostic confidence level (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — comment-context matches receive a ×1.5 weight because LLMs systematically over-annotate. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier as dense clusters constitute far stronger evidence of synthetic authorship than isolated hits.

Hyper-Verbose Identifiers578 hits · 486 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	248	def get_stable_macros_and_flags():	CODE
LOW	setup.py	449	def make_image_stable_extension():	CODE
LOW	test/smoke_test.py	24	def smoke_test_torchvision_read_decode() -> None:	CODE
LOW	test/smoke_test.py	69	def smoke_test_torchvision_decode_jpeg(device: str = "cpu"):	CODE
LOW	test/test_utils.py	42	def test_make_grid_not_inplace():	CODE
LOW	test/test_utils.py	56	def test_normalize_in_make_grid():	CODE
LOW	test/test_utils.py	83	def test_save_image_single_pixel():	CODE
LOW	test/test_utils.py	91	def test_save_image_file_object():	CODE
LOW	test/test_utils.py	103	def test_save_image_single_pixel_file_object():	CODE
LOW	test/test_utils.py	138	def test_draw_boxes_with_coloured_labels():	CODE
LOW	test/test_utils.py	153	def test_draw_boxes_with_coloured_label_backgrounds():	CODE
LOW	test/test_utils.py	170	def test_draw_boxes_with_coloured_label_text_boxes():	CODE
LOW	test/test_utils.py	208	def test_draw_rotated_boxes_fill():	CODE
LOW	test/test_utils.py	262	def test_draw_boxes_grayscale():	CODE
LOW	test/test_utils.py	321	def test_draw_segmentation_masks(colors, alpha, device):	CODE
LOW	test/test_utils.py	363	def test_draw_segmentation_masks_dtypes():	CODE
LOW	test/test_utils.py	384	def test_draw_segmentation_masks_errors(device):	CODE
LOW	test/test_utils.py	421	def test_draw_no_segmention_mask(device):	CODE
LOW	test/test_utils.py	430	def test_draw_keypoints_vanilla():	CODE
LOW	test/test_utils.py	465	def test_draw_keypoints_colored(colors):	CODE
LOW	test/test_utils.py	492	def test_draw_keypoints_visibility(connectivity, vis):	CODE
LOW	test/test_utils.py	531	def test_draw_keypoints_visibility_default():	CODE
LOW	test/test_utils.py	555	def test_draw_keypoints_dtypes():	CODE
LOW	test/test_utils.py	571	def test_draw_keypoints_errors():	CODE
LOW	test/test_utils.py	636	def test_flow_to_image_errors(input_flow, match):	CODE
LOW	test/conftest.py	25	def pytest_collection_modifyitems(items):	CODE
LOW	test/test_transforms_video.py	40	def test_random_resized_crop_video(self):	CODE
LOW	test/test_transforms_video.py	116	def samples_from_standard_normal(tensor):	CODE
LOW	test/test_transforms_video.py	164	def test_random_horizontal_flip_video(self, p):	CODE
LOW	test/test_datasets_utils.py	49	def test_get_redirect_url_max_hops_exceeded(self, mocker):	CODE
LOW	test/test_datasets_utils.py	81	def test_get_google_drive_file_id(self):	CODE
LOW	test/test_datasets_utils.py	88	def test_get_google_drive_file_id_invalid_url(self):	CODE
LOW	test/test_datasets_utils.py	116	def test_detect_file_type_incompatible(self, file):	CODE
LOW	test/test_datasets_utils.py	145	def test_decompress_no_compression(self):	CODE
LOW	test/test_datasets_utils.py	150	def test_decompress_remove_finished(self, tmpdir, use_pathlib):	CODE
LOW	test/test_datasets_utils.py	178	def test_extract_archive_defer_to_decompress(self, extension, remove_finished, mocker):	CODE
LOW	test/test_datasets_utils.py	270	def test_make_dataset_no_valid_files(tmpdir, kwargs, expected_error_msg):	CODE
LOW	test/test_datasets.py	149	def _create_annotation_folder(self, root, name, file_name_fn, num_examples):	CODE
LOW	test/test_datasets.py	385	def test_feature_types_target_color(self):	CODE
LOW	test/test_datasets.py	391	def test_feature_types_target_polygon(self):	CODE
LOW	test/test_datasets.py	1053	def test_not_found_or_corrupted(self):	CODE
LOW	test/test_datasets.py	1330	def test_not_found_or_corrupted(self):	CODE
LOW	test/test_datasets.py	1700	def _is_valid_file_to_extensions(self, is_valid_file):	CODE
LOW	test/test_datasets.py	1992	def test_tv_decode_image_support(self):	CODE
LOW	test/test_datasets.py	2727	def _meta_to_split_and_classification_ann(self, meta, idx):	CODE
LOW	test/test_datasets.py	3403	def _create_annotation_folder(num_examples: int, root_dir: str):	CODE
LOW	test/test_datasets.py	3511	def test_augmented_view_usage(self):	CODE
LOW	test/test_datasets_video_utils.py	72	def test_video_clips_custom_fps(self, tmpdir):	CODE
LOW	test/test_datasets_video_utils.py	83	def test_compute_clips_for_video(self):	CODE
LOW	test/test_backbone_utils.py	98	def _create_feature_extractor(self, args, *kwargs):	CODE
LOW	test/test_backbone_utils.py	131	def test_build_fx_feature_extractor(self, model_name):	CODE
LOW	test/test_backbone_utils.py	162	def test_node_name_conventions(self):	CODE
LOW	test/test_backbone_utils.py	186	def test_feature_extraction_methods_equivalence(self):	CODE
LOW	test/test_backbone_utils.py	207	def test_jit_forward_backward(self, model_name):	CODE
LOW	test/test_backbone_utils.py	294	def test_leaf_module_and_function(self):	CODE
LOW	test/test_transforms_tensor.py	31	def _test_transform_vs_scripted(transform, s_transform, tensor, msg=None):	CODE
LOW	test/test_transforms_tensor.py	39	def _test_transform_vs_scripted_on_batch(transform, s_transform, batch_tensors, msg=None):	CODE
LOW	test/test_transforms_tensor.py	140	def test_color_jitter_brightness(self, brightness, device, channels):	CODE
LOW	test/test_transforms_tensor.py	154	def test_color_jitter_contrast(self, contrast, device, channels):	CODE
LOW	test/test_transforms_tensor.py	168	def test_color_jitter_saturation(self, saturation, device, channels):	CODE
518 more matches not shown…

Unused Imports587 hits · 403 pts

Severity	File	Line	Context
LOW	hubconf.py	4	CODE
LOW	hubconf.py	4	CODE
LOW	hubconf.py	5	CODE
LOW	hubconf.py	6	CODE
LOW	hubconf.py	6	CODE
LOW	hubconf.py	6	CODE
LOW	hubconf.py	6	CODE
LOW	hubconf.py	7	CODE
LOW	hubconf.py	7	CODE
LOW	hubconf.py	7	CODE
LOW	hubconf.py	7	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	8	CODE
LOW	hubconf.py	21	CODE
LOW	hubconf.py	22	CODE
LOW	hubconf.py	23	CODE
LOW	hubconf.py	24	CODE
LOW	hubconf.py	24	CODE
LOW	hubconf.py	24	CODE
LOW	hubconf.py	24	CODE
LOW	hubconf.py	25	CODE
LOW	hubconf.py	26	CODE
LOW	hubconf.py	26	CODE
LOW	hubconf.py	27	CODE
LOW	hubconf.py	27	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	28	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	45	CODE
LOW	hubconf.py	57	CODE
527 more matches not shown…

Cross-File Repetition66 hits · 330 pts

Severity	File	Line	Snippet	Context
HIGH	references/classification/utils.py	0	track a series of values and provide access to smoothed values over a window or the global series average.	STRING
HIGH	references/depth/stereo/utils/logger.py	0	track a series of values and provide access to smoothed values over a window or the global series average.	STRING
HIGH	references/segmentation/utils.py	0	track a series of values and provide access to smoothed values over a window or the global series average.	STRING
HIGH	references/video_classification/utils.py	0	track a series of values and provide access to smoothed values over a window or the global series average.	STRING
HIGH	references/optical_flow/utils.py	0	track a series of values and provide access to smoothed values over a window or the global series average.	STRING
HIGH	references/detection/utils.py	0	track a series of values and provide access to smoothed values over a window or the global series average.	STRING
HIGH	references/classification/utils.py	0	this function disables printing when not in master process	STRING
HIGH	references/segmentation/utils.py	0	this function disables printing when not in master process	STRING
HIGH	references/video_classification/utils.py	0	this function disables printing when not in master process	STRING
HIGH	references/detection/utils.py	0	this function disables printing when not in master process	STRING
HIGH	torchvision/datasets/stl10.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/usps.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/svhn.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/semeion.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/cifar.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/caltech.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/mnist.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/folder.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/fakedata.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/datasets/omniglot.py	0	args: index (int): index returns: tuple: (image, target) where target is index of the target character class.	STRING
HIGH	torchvision/models/shufflenetv2.py	0	these weights were trained from scratch by using torchvision's `new training recipe <https://pytorch.org/blog/how-to-tra	STRING
HIGH	torchvision/models/mnasnet.py	0	these weights were trained from scratch by using torchvision's `new training recipe <https://pytorch.org/blog/how-to-tra	STRING
HIGH	torchvision/models/resnet.py	0	these weights were trained from scratch by using torchvision's `new training recipe <https://pytorch.org/blog/how-to-tra	STRING
HIGH	torchvision/models/regnet.py	0	these weights reproduce closely the results of the paper using a simplified training recipe.	STRING
HIGH	torchvision/models/squeezenet.py	0	these weights reproduce closely the results of the paper using a simplified training recipe.	STRING
HIGH	torchvision/models/resnet.py	0	these weights reproduce closely the results of the paper using a simplified training recipe.	STRING
HIGH	torchvision/models/mobilenetv2.py	0	these weights reproduce closely the results of the paper using a simplified training recipe.	STRING
HIGH	torchvision/models/swin_transformer.py	0	these weights reproduce closely the results of the paper using a simplified training recipe.	STRING
HIGH	torchvision/models/alexnet.py	0	these weights reproduce closely the results of the paper using a simplified training recipe.	STRING
HIGH	torchvision/models/regnet.py	0	these weights improve marginally upon the results of the original paper by using a modified version of torchvision's `ne	STRING
HIGH	torchvision/models/efficientnet.py	0	these weights improve marginally upon the results of the original paper by using a modified version of torchvision's `ne	STRING
HIGH	torchvision/models/mobilenetv2.py	0	these weights improve marginally upon the results of the original paper by using a modified version of torchvision's `ne	STRING
HIGH	torchvision/models/convnext.py	0	these weights improve marginally upon the results of the original paper by using a modified version of torchvision's `ne	STRING
HIGH	torchvision/models/mobilenetv3.py	0	these weights improve marginally upon the results of the original paper by using a modified version of torchvision's `ne	STRING
HIGH	torchvision/models/segmentation/fcn.py	0	these weights were trained on a subset of coco, using only the 20 categories that are present in the pascal voc dataset.	STRING
HIGH	torchvision/models/segmentation/lraspp.py	0	these weights were trained on a subset of coco, using only the 20 categories that are present in the pascal voc dataset.	STRING
HIGH	torchvision/models/segmentation/deeplabv3.py	0	these weights were trained on a subset of coco, using only the 20 categories that are present in the pascal voc dataset.	STRING
HIGH	torchvision/models/quantization/shufflenetv2.py	0	fuse conv/bn/relu modules in resnet models fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization. mod	STRING
HIGH	torchvision/models/quantization/googlenet.py	0	fuse conv/bn/relu modules in resnet models fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization. mod	STRING
HIGH	torchvision/models/quantization/inception.py	0	fuse conv/bn/relu modules in resnet models fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization. mod	STRING
HIGH	torchvision/models/quantization/resnet.py	0	fuse conv/bn/relu modules in resnet models fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization. mod	STRING
HIGH	torchvision/models/quantization/shufflenetv2.py	0	these weights were produced by doing quantization aware training (eager mode) on top of the unquantized weights listed b	STRING
HIGH	torchvision/models/quantization/googlenet.py	0	these weights were produced by doing quantization aware training (eager mode) on top of the unquantized weights listed b	STRING
HIGH	torchvision/models/quantization/inception.py	0	these weights were produced by doing quantization aware training (eager mode) on top of the unquantized weights listed b	STRING
HIGH	torchvision/models/quantization/resnet.py	0	these weights were produced by doing quantization aware training (eager mode) on top of the unquantized weights listed b	STRING
HIGH	torchvision/models/quantization/mobilenetv2.py	0	these weights were produced by doing quantization aware training (eager mode) on top of the unquantized weights listed b	STRING
HIGH	torchvision/models/quantization/mobilenetv3.py	0	these weights were produced by doing quantization aware training (eager mode) on top of the unquantized weights listed b	STRING
HIGH	torchvision/models/detection/ssdlite.py	0	these weights were produced by following a similar training recipe as on the paper.	STRING
HIGH	torchvision/models/detection/ssd.py	0	these weights were produced by following a similar training recipe as on the paper.	STRING
HIGH	torchvision/models/detection/faster_rcnn.py	0	these weights were produced by following a similar training recipe as on the paper.	STRING
HIGH	torchvision/models/detection/fcos.py	0	these weights were produced by following a similar training recipe as on the paper.	STRING
HIGH	torchvision/models/detection/mask_rcnn.py	0	these weights were produced by following a similar training recipe as on the paper.	STRING
HIGH	torchvision/models/detection/retinanet.py	0	these weights were produced by following a similar training recipe as on the paper.	STRING
HIGH	torchvision/models/detection/keypoint_rcnn.py	0	these weights were produced by following a similar training recipe as on the paper.	STRING
HIGH	torchvision/models/detection/faster_rcnn.py	0	implements keypoint r-cnn. the input to the model is expected to be a list of tensors, each of shape [c, h, w], one for	STRING
HIGH	torchvision/models/detection/mask_rcnn.py	0	implements keypoint r-cnn. the input to the model is expected to be a list of tensors, each of shape [c, h, w], one for	STRING
HIGH	torchvision/models/detection/keypoint_rcnn.py	0	implements keypoint r-cnn. the input to the model is expected to be a list of tensors, each of shape [c, h, w], one for	STRING
HIGH	torchvision/models/detection/faster_rcnn.py	0	these weights were produced using an enhanced training recipe to boost the model accuracy.	STRING
HIGH	torchvision/models/detection/mask_rcnn.py	0	these weights were produced using an enhanced training recipe to boost the model accuracy.	STRING
HIGH	torchvision/models/detection/retinanet.py	0	these weights were produced using an enhanced training recipe to boost the model accuracy.	STRING
6 more matches not shown…

Decorative Section Separators49 hits · 147 pts

Severity	File	Line	Snippet	Context
MEDIUM	gallery/others/plot_optical_flow.py	51	# --------------------------------	COMMENT
MEDIUM	gallery/others/plot_optical_flow.py	112	# ----------------------------------	COMMENT
MEDIUM	gallery/others/plot_optical_flow.py	152	# ---------------------------	COMMENT
MEDIUM	gallery/others/plot_optical_flow.py	174	# ---------------------------------------	COMMENT
MEDIUM	gallery/others/plot_visualization_utils.py	39	# ----------------------------	COMMENT
MEDIUM	gallery/others/plot_visualization_utils.py	57	# --------------------------	COMMENT
MEDIUM	gallery/others/plot_visualization_utils.py	105	# ------------------------------	COMMENT
MEDIUM	gallery/others/plot_visualization_utils.py	357	# ------------------------------	COMMENT
MEDIUM	gallery/others/plot_repurposing_annotations.py	62	# -----------------------------------------------	COMMENT
MEDIUM	gallery/others/plot_repurposing_annotations.py	162	# ----------------------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_cutmix_mixup.py	32	# -----------------------	COMMENT
MEDIUM	gallery/transforms/plot_cutmix_mixup.py	69	# -----------------------------	COMMENT
MEDIUM	gallery/transforms/plot_cutmix_mixup.py	125	# -------------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_getting_started.py	36	# ----------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_getting_started.py	49	# --------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_getting_started.py	83	# -------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_getting_started.py	128	# --------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_getting_started.py	156	# ------------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_getting_started.py	225	# ------------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_custom_transforms.py	24	# ===============================================================	COMMENT
MEDIUM	gallery/transforms/plot_custom_transforms.py	84	# =====================================	COMMENT
MEDIUM	gallery/transforms/plot_custom_transforms.py	113	# -----------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_custom_transforms.py	151	# --------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_keypoints_transforms.py	43	# -------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_keypoints_transforms.py	75	# ------------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_illustrations.py	37	# --------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_illustrations.py	141	# ----------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_illustrations.py	253	# -----------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_illustrations.py	299	# ---------------------------	COMMENT
MEDIUM	gallery/transforms/plot_rotated_box_transforms.py	41	# -------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_rotated_box_transforms.py	64	# ------------------------	COMMENT
MEDIUM	gallery/transforms/plot_rotated_box_transforms.py	97	# -------------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_rotated_box_transforms.py	169	# --------------------------	COMMENT
MEDIUM	gallery/transforms/plot_tv_tensors.py	35	# -------------------	COMMENT
MEDIUM	gallery/transforms/plot_tv_tensors.py	58	# ------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_tv_tensors.py	69	# ------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_tv_tensors.py	137	# -----------------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_e2e.py	43	# -------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_e2e.py	93	# ----------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_e2e.py	142	# ------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_transforms_e2e.py	172	# -------------------	COMMENT
MEDIUM	gallery/transforms/plot_custom_tv_tensors.py	93	# -----------------------------------------------------------------------	COMMENT
MEDIUM	gallery/transforms/plot_custom_tv_tensors.py	125	# ------------------------------	COMMENT
MEDIUM	torchvision/_autograd_registrations.py	8	# =====================================================================	COMMENT
MEDIUM	torchvision/_autograd_registrations.py	10	# =====================================================================	COMMENT
MEDIUM	torchvision/_autograd_registrations.py	208	# =====================================================================	COMMENT
MEDIUM	torchvision/_autograd_registrations.py	210	# =====================================================================	COMMENT
MEDIUM	torchvision/_autograd_registrations.py	328	# =====================================================================	COMMENT
MEDIUM	torchvision/_autograd_registrations.py	330	# =====================================================================	COMMENT

Over-Commented Block117 hits · 116 pts

Severity	File	Line	Snippet	Context
LOW	packaging/cut_release.sh	1	#!/usr/bin/env bash	COMMENT
LOW	test/smoke_test.py	41	# if torch.cuda.is_available():	COMMENT
LOW	test/conftest.py	21	config.addinivalue_line("markers", "dont_collect: mark for tests that should not be collected")	COMMENT
LOW	test/conftest.py	81		COMMENT
LOW	test/test_image.py	861	@pytest.mark.parametrize("scripted", (True, False))	COMMENT
LOW	test/test_models.py	841	# We first try to assert the entire output if possible. This is not	COMMENT
LOW	test/test_transforms_v2.py	7161	dtype=torch.float32,	COMMENT
LOW	references/depth/stereo/transforms.py	481	)	COMMENT
LOW	gallery/others/plot_optical_flow.py	121	# If you can, run this example on a GPU, it will be a lot faster.	COMMENT
LOW	gallery/others/plot_optical_flow.py	141	# vertical displacement of each pixel from the first image to the second image.	COMMENT
LOW	gallery/others/plot_optical_flow.py	181		COMMENT
LOW	gallery/others/plot_visualization_utils.py	101	show(dogs_with_boxes)	COMMENT
LOW	gallery/others/plot_visualization_utils.py	161	# 1]``. To get boolean masks, we can do the following:	COMMENT
LOW	gallery/others/plot_visualization_utils.py	241	# Instance segmentation models	COMMENT
LOW	gallery/others/plot_visualization_utils.py	281	# associated to those keys has ``num_instances`` elements in it. In our case	COMMENT
LOW	gallery/others/plot_visualization_utils.py	341		COMMENT
LOW	gallery/others/plot_visualization_utils.py	381	# As we see the output contains a list of dictionaries.	COMMENT
LOW	gallery/others/plot_visualization_utils.py	421	# As we see, the keypoints appear as colored circles over the image.	COMMENT
LOW	gallery/others/plot_visualization_utils.py	441	#	COMMENT
LOW	gallery/others/plot_visualization_utils.py	501	# So we definitely don't want to display those keypoints and connections, and you don't have to.	COMMENT
LOW	gallery/others/plot_visualization_utils.py	521	# The :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn` model,	COMMENT
LOW	gallery/others/plot_scripted_tensor_transforms.py	61	# <v1_or_v2>` way to use transforms in your code.	COMMENT
LOW	gallery/others/plot_repurposing_annotations.py	41		COMMENT
LOW	gallery/others/plot_repurposing_annotations.py	161	# Converting Segmentation Dataset to Detection Dataset	COMMENT
LOW	gallery/transforms/plot_cutmix_mixup.py	61	print(f"{images.shape = }, {labels.shape = }")	COMMENT
LOW	gallery/transforms/plot_cutmix_mixup.py	81	mixup = v2.MixUp(num_classes=NUM_CLASSES)	COMMENT
LOW	gallery/transforms/plot_cutmix_mixup.py	121	break	COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	61		COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	121	# answer these in the next sections.	COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	141	img_dp = tv_tensors.Image(torch.randint(0, 256, (3, 256, 256), dtype=torch.uint8))	COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	181	# %%	COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	201	# :class:`~torchvision.tv_tensors.Image` class at all, and yet our images	COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	221	#	COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	241	# TVTensors, so they don't return TVTensors out of the box.	COMMENT
LOW	gallery/transforms/plot_transforms_getting_started.py	261	#	COMMENT
LOW	gallery/transforms/plot_custom_transforms.py	21		COMMENT
LOW	gallery/transforms/plot_custom_transforms.py	81	# :ref:`tv_tensor_unwrapping_behaviour`	COMMENT
LOW	gallery/transforms/plot_custom_transforms.py	141	# ``structured_input``, the input is flattened and then each individual part is	COMMENT
LOW	gallery/transforms/plot_custom_transforms.py	181	_ = my_random_transform(structured_input) # doesn't transform	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	41	#	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	61	# ~~~~~~~~~~	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	81	# (see also :func:`~torchvision.transforms.functional.perspective`)	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	101	# (see also :func:`~torchvision.transforms.functional.affine`)	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	121	# The :class:`~torchvision.transforms.RandomCrop` transform	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	141	# ----------------------	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	161	# ~~~~~~~~~~~	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	181	# The :class:`~torchvision.transforms.RandomInvert` transform	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	201	# ~~~~~~~~~~~~~~	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	221	# RandomAutocontrast	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	241	# JPEG	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	281	# ~~~~~~~~~~~~~~~~~~	COMMENT
LOW	gallery/transforms/plot_transforms_illustrations.py	301	# The following transforms are randomly-applied given a probability ``p``. That is, given ``p = 0.5``,	COMMENT
LOW	gallery/transforms/plot_rotated_box_transforms.py	81	resized_imgs = [	COMMENT
LOW	gallery/transforms/plot_rotated_box_transforms.py	101	# transformed (cropped) image. In general, this may happen on most of the	COMMENT
LOW	gallery/transforms/plot_rotated_box_transforms.py	161	print(soft_center_crops_and_boxes)	COMMENT
LOW	gallery/transforms/plot_tv_tensors.py	41		COMMENT
LOW	gallery/transforms/plot_tv_tensors.py	61	# Everything that is supported on a plain :class:`torch.Tensor` like ``.sum()`` or	COMMENT
LOW	gallery/transforms/plot_tv_tensors.py	81	# Similar to other PyTorch creations ops, the constructor also takes the ``dtype``, ``device``, and ``requires_grad``	COMMENT
LOW	gallery/transforms/plot_tv_tensors.py	121	# into a TVTensor. This is useful when you already have an object of the	COMMENT
LOW	gallery/transforms/plot_tv_tensors.py	181	# Why is this happening?	COMMENT
57 more matches not shown…

Deep Nesting96 hits · 91 pts

Severity	File	Line	Context
LOW	setup.py	522	CODE
LOW	packaging/wheel/relocate.py	282	CODE
LOW	test/conftest.py	25	CODE
LOW	test/test_datasets.py	292	CODE
LOW	test/test_datasets.py	1002	CODE
LOW	test/test_datasets.py	1763	CODE
LOW	test/test_datasets.py	2530	CODE
LOW	test/test_datasets_download.py	80	CODE
LOW	test/test_extended_models.py	244	CODE
LOW	test/test_transforms_v2.py	7194	CODE
LOW	test/test_transforms_v2.py	1789	CODE
LOW	test/test_transforms_v2.py	4238	CODE
LOW	test/test_ops.py	290	CODE
LOW	test/test_ops.py	340	CODE
LOW	test/test_ops.py	427	CODE
LOW	test/test_ops.py	661	CODE
LOW	test/test_ops.py	962	CODE
LOW	test/common_utils.py	410	CODE
LOW	test/common_utils.py	54	CODE
LOW	test/datasets_utils.py	971	CODE
LOW	test/datasets_utils.py	483	CODE
LOW	test/datasets_utils.py	576	CODE
LOW	test/datasets_utils.py	709	CODE
LOW	references/classification/presets.py	21	CODE
LOW	references/depth/stereo/transforms.py	519	CODE
LOW	references/depth/stereo/train.py	36	CODE
LOW	references/depth/stereo/train.py	267	CODE
LOW	references/depth/stereo/cascade_evaluation.py	109	CODE
LOW	references/optical_flow/train.py	13	CODE
LOW	references/optical_flow/train.py	133	CODE
LOW	references/detection/presets.py	21	CODE
LOW	gallery/transforms/helpers.py	9	CODE
LOW	torchvision/utils.py	25	CODE
LOW	torchvision/utils.py	177	CODE
LOW	torchvision/utils.py	285	CODE
LOW	torchvision/utils.py	418	CODE
LOW	torchvision/utils.py	487	CODE
LOW	torchvision/utils.py	722	CODE
LOW	torchvision/datasets/flickr.py	40	CODE
LOW	torchvision/datasets/folder.py	49	CODE
LOW	torchvision/datasets/lfw.py	137	CODE
LOW	torchvision/datasets/celeba.py	175	CODE
LOW	torchvision/datasets/_stereo_matching.py	579	CODE
LOW	torchvision/datasets/widerface.py	113	CODE
LOW	torchvision/datasets/_optical_flow.py	349	CODE
LOW	torchvision/datasets/inaturalist.py	142	CODE
LOW	torchvision/datasets/inaturalist.py	209	CODE
LOW	torchvision/models/googlenet.py	32	CODE
LOW	torchvision/models/mnasnet.py	112	CODE
LOW	torchvision/models/vgg.py	36	CODE
LOW	torchvision/models/regnet.py	297	CODE
LOW	torchvision/models/vision_transformer.py	789	CODE
LOW	torchvision/models/vision_transformer.py	54	CODE
LOW	torchvision/models/densenet.py	152	CODE
LOW	torchvision/models/inception.py	29	CODE
LOW	torchvision/models/feature_extraction.py	369	CODE
LOW	torchvision/models/efficientnet.py	366	CODE
LOW	torchvision/models/efficientnet.py	234	CODE
LOW	torchvision/models/resnet.py	167	CODE
LOW	torchvision/models/mobilenetv2.py	68	CODE
36 more matches not shown…

Docstring Block Structure12 hits · 60 pts

Severity	File	Line	Snippet	Context
HIGH	test/datasets_utils.py	881	Create a video file from random data. Args: root (Union[str, pathlib.Path]): Root directory the video file	STRING
HIGH	test/datasets_utils.py	929	Create a folder of random videos. Args: root (Union[str, pathlib.Path]): Root directory the video folder wi	STRING
HIGH	references/classification/utils.py	324	This method can be used to prepare weights files for new models. It receives as input a model architecture and	STRING
HIGH	torchvision/utils.py	728	Parses a specification of colors for a set of objects. Args: colors: A specification of colors for the	STRING
HIGH	torchvision/datasets/places365.py	13	`Places365 <http://places2.csail.mit.edu/index.html>`_ classification dataset. Args: root (str or ``pathlib	STRING
HIGH	torchvision/datasets/folder.py	174	Generates a list of samples of a form (path_to_sample, class). This can be overridden to e.g. read files from a	STRING
HIGH	torchvision/datasets/folder.py	208	Find the class folders in a dataset structured as follows:: directory/ ├── class_x	STRING
HIGH	torchvision/datasets/utils.py	248	Detect the archive type and/or compression of a file. Args: file (str): the filename Returns:	STRING
HIGH	torchvision/datasets/kinetics.py	23	`Generic Kinetics <https://www.deepmind.com/open-source/kinetics>`_ dataset. Kinetics-400/600/700 are action re	STRING
HIGH	torchvision/models/feature_extraction.py	210	Dev utility to return node names in order of execution. See note on node names under :func:`create_feature_extr	STRING
HIGH	torchvision/transforms/functional.py	217	Convert a tensor image to the given ``dtype`` and scale the values accordingly This function does not support PIL Im	STRING
HIGH	torchvision/ops/deform_conv.py	24	Performs Deformable Convolution v2, described in `Deformable ConvNets v2: More Deformable, Better Results <	STRING

Modern Structural Boilerplate53 hits · 53 pts

Severity	File	Line	Snippet	Context
LOW	test/datasets_utils.py	38	__all__ = [	CODE
LOW	references/depth/stereo/utils/losses.py	95	def set_gamma(self, gamma: float) -> None:	CODE
LOW	references/depth/stereo/utils/losses.py	373	def set_gamma(self, gamma: float) -> None:	CODE
LOW	torchvision/utils.py	13	__all__ = [	CODE
LOW	torchvision/datasets/samplers/clip_sampler.py	102	def set_epoch(self, epoch: int) -> None:	CODE
LOW	torchvision/io/__init__.py	49	__all__ = [	CODE
LOW	torchvision/io/video.py	14	__all__ = [	CODE
LOW	torchvision/models/shufflenetv2.py	15	__all__ = [	CODE
LOW	torchvision/models/googlenet.py	18	__all__ = ["GoogLeNet", "GoogLeNetOutputs", "_GoogLeNetOutputs", "GoogLeNet_Weights", "googlenet"]	CODE
LOW	torchvision/models/mnasnet.py	16	__all__ = [	CODE
LOW	torchvision/models/vgg.py	14	__all__ = [	CODE
LOW	torchvision/models/regnet.py	17	__all__ = [	CODE
LOW	torchvision/models/vision_transformer.py	17	__all__ = [	CODE
LOW	torchvision/models/squeezenet.py	15	__all__ = ["SqueezeNet", "SqueezeNet1_0_Weights", "SqueezeNet1_1_Weights", "squeezenet1_0", "squeezenet1_1"]	CODE
LOW	torchvision/models/_api.py	18	__all__ = ["WeightsEnum", "Weights", "get_model", "get_model_builder", "get_model_weights", "get_weight", "list_models"]	CODE
LOW	torchvision/models/densenet.py	18	__all__ = [	CODE
LOW	torchvision/models/inception.py	17	__all__ = ["Inception3", "InceptionOutputs", "_InceptionOutputs", "Inception_V3_Weights", "inception_v3"]	CODE
LOW	torchvision/models/feature_extraction.py	17	__all__ = ["create_feature_extractor", "get_graph_node_names"]	CODE
LOW	torchvision/models/efficientnet.py	20	__all__ = [	CODE
LOW	torchvision/models/resnet.py	15	__all__ = [	CODE
LOW	torchvision/models/mobilenetv2.py	15	__all__ = ["MobileNetV2", "MobileNet_V2_Weights", "mobilenet_v2"]	CODE
LOW	torchvision/models/mobilenetv3.py	16	__all__ = [	CODE
LOW	torchvision/models/swin_transformer.py	18	__all__ = [	CODE
LOW	torchvision/models/alexnet.py	14	__all__ = ["AlexNet", "AlexNet_Weights", "alexnet"]	CODE
LOW	torchvision/models/maxvit.py	19	__all__ = [	CODE
LOW	torchvision/models/convnext.py	18	__all__ = [	CODE
LOW	torchvision/models/video/s3d.py	15	__all__ = [	CODE
LOW	torchvision/models/video/resnet.py	15	__all__ = [	CODE
LOW	torchvision/models/video/mvit.py	19	__all__ = [	CODE
LOW	torchvision/models/video/swin_transformer.py	21	__all__ = [	CODE
LOW	torchvision/models/segmentation/fcn.py	14	__all__ = ["FCN", "FCN_ResNet50_Weights", "FCN_ResNet101_Weights", "fcn_resnet50", "fcn_resnet101"]	CODE
LOW	torchvision/models/segmentation/lraspp.py	16	__all__ = ["LRASPP", "LRASPP_MobileNet_V3_Large_Weights", "lraspp_mobilenet_v3_large"]	CODE
LOW	torchvision/models/segmentation/deeplabv3.py	19	__all__ = [	CODE
LOW	torchvision/models/quantization/shufflenetv2.py	22	__all__ = [	CODE
LOW	torchvision/models/quantization/googlenet.py	18	__all__ = [	CODE
LOW	torchvision/models/quantization/inception.py	19	__all__ = [	CODE
LOW	torchvision/models/quantization/resnet.py	24	__all__ = [	CODE
LOW	torchvision/models/quantization/mobilenetv2.py	16	__all__ = [	CODE
LOW	torchvision/models/quantization/mobilenetv3.py	23	__all__ = [	CODE
LOW	torchvision/models/detection/ssdlite.py	23	__all__ = [	CODE
LOW	torchvision/models/detection/ssd.py	22	__all__ = [	CODE
LOW	torchvision/models/detection/faster_rcnn.py	24	__all__ = [	CODE
LOW	torchvision/models/detection/fcos.py	24	__all__ = [	CODE
LOW	torchvision/models/detection/mask_rcnn.py	18	__all__ = [	CODE
LOW	torchvision/models/detection/retinanet.py	25	__all__ = [	CODE
LOW	torchvision/models/detection/keypoint_rcnn.py	18	__all__ = [	CODE
LOW	torchvision/models/detection/_utils.py	397	def set_low_quality_matches_(self, matches: Tensor, all_matches: Tensor, match_quality_matrix: Tensor) -> None:	CODE
LOW	torchvision/transforms/transforms.py	20	__all__ = [	CODE
LOW	torchvision/transforms/_presets.py	14	__all__ = [	CODE
LOW	torchvision/transforms/autoaugment.py	10	__all__ = ["AutoAugmentPolicy", "AutoAugment", "RandAugment", "TrivialAugmentWide", "AugMix"]	CODE
LOW	torchvision/transforms/_transforms_video.py	12	__all__ = [	CODE
LOW	torchvision/tv_tensors/_dataset_wrapper.py	16	__all__ = ["wrap_dataset_for_transforms_v2"]	CODE
LOW	torchvision/ops/__init__.py	33	__all__ = [	CODE

AI Structural Patterns43 hits · 43 pts

Severity	File	Line	Context
LOW	test/common_extended_utils.py	186	CODE
LOW	test/common_utils.py	313	CODE
LOW	references/classification/presets.py	21	CODE
LOW	references/depth/stereo/presets.py	45	CODE
LOW	references/depth/stereo/cascade_evaluation.py	109	CODE
LOW	references/optical_flow/presets.py	23	CODE
LOW	torchvision/utils.py	285	CODE
LOW	torchvision/datasets/video_utils.py	102	CODE
LOW	torchvision/datasets/ucf101.py	54	CODE
LOW	torchvision/datasets/hmdb51.py	61	CODE
LOW	torchvision/datasets/kinetics.py	91	CODE
LOW	torchvision/io/image.py	77	CODE
LOW	torchvision/io/image.py	131	CODE
LOW	torchvision/io/image.py	153	CODE
LOW	torchvision/io/image.py	351	CODE
LOW	torchvision/models/shufflenetv2.py	101	CODE
LOW	torchvision/models/densenet.py	218	CODE
LOW	torchvision/models/resnet.py	105	CODE
LOW	torchvision/models/resnet.py	163	CODE
LOW	torchvision/models/swin_transformer.py	528	CODE
LOW	torchvision/models/maxvit.py	211	CODE
LOW	torchvision/models/maxvit.py	224	CODE
LOW	torchvision/models/maxvit.py	590	CODE
LOW	torchvision/models/video/s3d.py	89	CODE
LOW	torchvision/models/video/resnet.py	121	CODE
LOW	torchvision/models/video/resnet.py	170	CODE
LOW	torchvision/models/video/mvit.py	426	CODE
LOW	torchvision/models/video/swin_transformer.py	386	CODE
LOW	torchvision/models/quantization/shufflenetv2.py	49	CODE
LOW	torchvision/models/quantization/resnet.py	57	CODE
LOW	torchvision/models/quantization/resnet.py	88	CODE
LOW	torchvision/models/detection/ssd.py	186	CODE
LOW	torchvision/models/detection/faster_rcnn.py	167	CODE
LOW	torchvision/models/detection/fcos.py	365	CODE
LOW	torchvision/models/detection/mask_rcnn.py	165	CODE
LOW	torchvision/models/detection/retinanet.py	417	CODE
LOW	torchvision/models/detection/keypoint_rcnn.py	164	CODE
LOW	torchvision/transforms/_functional_tensor.py	255	CODE
LOW	torchvision/ops/misc.py	70	CODE
LOW	torchvision/ops/misc.py	145	CODE
LOW	torchvision/ops/misc.py	194	CODE
LOW	torchvision/ops/roi_align.py	200	CODE
LOW	torchvision/ops/feature_pyramid_network.py	204	CODE

AI Slop Vocabulary9 hits · 21 pts

Severity	File	Line	Snippet	Context
MEDIUM	test/conftest.py	63	# TODO: something more robust would be to do that only in a sandcastle instance,	COMMENT
MEDIUM	test/test_models.py	253	# This may be caused by the harness environment (e.g. num classes, input initialization	COMMENT
LOW	test/test_transforms_v2.py	155	# Most kernels just return a tensor, but some also return some additional metadata	COMMENT
LOW	test/test_transforms_v2.py	174	# Most kernels just return a tensor, but some also return some additional metadata	COMMENT
LOW	references/detection/transforms.py	464	# Let's just return (image, target)	COMMENT
MEDIUM	gallery/transforms/plot_transforms_e2e.py	99	# dataset so that they can leverage multi-processing from the	COMMENT
MEDIUM	torchvision/datasets/caltech.py	126	# can be more robust and check hash of files	COMMENT
MEDIUM	torchvision/datasets/caltech.py	226	# can be more robust and check hash of files	COMMENT
LOW	torchvision/transforms/v2/_transform.py	155	# are JIT scriptable, and we made sure that for single image inputs v1 and v2 are equivalent, we just return the	COMMENT

Self-Referential Comments6 hits · 18 pts

Severity	File	Line	Snippet	Context
MEDIUM	gallery/transforms/plot_rotated_box_transforms.py	40	# Creating a Rotated Bounding Box	COMMENT
MEDIUM	torchvision/transforms/v2/_transform.py	129	# This method is called after subclassing has happened, i.e. `cls` is the subclass, e.g. `Resize`.	COMMENT
MEDIUM	torchvision/transforms/v2/_transform.py	134	# This method is called by `__prepare_scriptable__` to instantiate the equivalent v1 transform from the current	COMMENT
MEDIUM	torchvision/transforms/v2/_transform.py	153	# This method is called early on when `torch.jit.script`'ing an `nn.Module` instance. If it succeeds, the return	COMMENT
MEDIUM	torchvision/transforms/v2/functional/_color.py	263	# Create a view on the underlying output while pointing at the same data. We do this to avoid indexing twice.	COMMENT
MEDIUM	docs/source/conf.py	6	# This file is execfile()d with the current directory set to its	COMMENT

Excessive Try-Catch Wrapping12 hits · 18 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	64	except Exception:	CODE
LOW	test/test_extended_models.py	398	except Exception:	CODE
LOW	test/test_models.py	992	except Exception as e:	CODE
MEDIUM	test/test_transforms_v2.py	105	def _script(obj):	CODE
LOW	test/test_transforms_v2.py	108	except Exception as error:	CODE
LOW	test/common_utils.py	249	except Exception as exc:	CODE
LOW	torchvision/utils.py	192	except Exception:	CODE
LOW	torchvision/datasets/video_utils.py	327	except Exception:	CODE
LOW⚡	release/apply-release-changes.py	46	except Exception as e:	CODE
MEDIUM⚡	release/apply-release-changes.py	47	print(f"Error processing {file_path}: {e}")	CODE
MEDIUM	release/apply-release-changes.py	71	print("Error: Exactly one version parameter is required")	CODE
MEDIUM	release/apply-release-changes.py	83	print("Error: Not inside a git repository. Please run from within a git repository.")	CODE

Verbosity Indicators9 hits · 16 pts

Severity	File	Line	Snippet	Context
LOW⚡	test/smoke_test.py	93	# Step 1: Initialize model with the best available weights	COMMENT
LOW⚡	test/smoke_test.py	98	# Step 2: Initialize the inference transforms	COMMENT
LOW⚡	test/smoke_test.py	101	# Step 3: Apply inference preprocessing transforms	COMMENT
LOW⚡	test/smoke_test.py	104	# Step 4: Use the model and print the predicted category	COMMENT
LOW	torchvision/csrc/ops/box_iou_rotated_utils.h	182	// Step 1:	COMMENT
LOW	torchvision/csrc/ops/box_iou_rotated_utils.h	194	// Step 2:	COMMENT
LOW	torchvision/csrc/ops/box_iou_rotated_utils.h	205	// Step 3:	COMMENT
LOW	torchvision/csrc/ops/box_iou_rotated_utils.h	262	// Step 4:	COMMENT
LOW	torchvision/csrc/ops/box_iou_rotated_utils.h	279	// Step 5:	COMMENT

Redundant / Tautological Comments9 hits · 15 pts

Severity	File	Line	Snippet	Context
LOW	packaging/pre_build_script_arm64.sh	25	# Set vcpkg to only build release packages	COMMENT
LOW	test/test_utils.py	132	# Check if modification is not in place	COMMENT
LOW	test/test_utils.py	257	# Check if modification is not in place	COMMENT
LOW	test/test_transforms.py	1743	# Check if content in center of both image and cropped output is same.	COMMENT
LOW⚡	references/similarity/loss.py	102	# Check if labels[i] == labels[j]	COMMENT
LOW	torchvision/models/detection/roi_heads.py	423	# Set shape to [batchxCxHxW]	COMMENT
LOW	torchvision/models/detection/roi_heads.py	449	# Set shape to [batchxCxHxW]	COMMENT
LOW⚡	release/apply-release-changes.py	56	# Check if .git directory exists	COMMENT
LOW	release/apply-release-changes.py	69	# Check if version is provided as command line argument	COMMENT

Structural Annotation Overuse6 hits · 12 pts

Severity	File	Line	Snippet	Context
LOW⚡	test/smoke_test.py	93	# Step 1: Initialize model with the best available weights	COMMENT
LOW⚡	test/smoke_test.py	98	# Step 2: Initialize the inference transforms	COMMENT
LOW⚡	test/smoke_test.py	101	# Step 3: Apply inference preprocessing transforms	COMMENT
LOW⚡	test/smoke_test.py	104	# Step 4: Use the model and print the predicted category	COMMENT
LOW	torchvision/transforms/v2/functional/_meta.py	276	# NOTE: This function cannot modify the input tensor inplace as it requires a dimension change.	COMMENT
LOW	torchvision/transforms/v2/functional/_meta.py	301	# NOTE: This function cannot modify the input tensor inplace as it requires a dimension change.	COMMENT

Slop Phrases5 hits · 8 pts

Severity	File	Line	Snippet	Context
LOW	references/depth/stereo/train.py	527	# make sure to disable shuffling since the dataset is already shuffled	COMMENT
LOW	gallery/transforms/plot_custom_transforms.py	79	# While working with TVTensor classes in your code, make sure to	COMMENT
MEDIUM	gallery/transforms/plot_tv_tensors.py	172	# Alternatively, you can use the :func:`~torchvision.tv_tensors.set_return_type`	COMMENT
LOW	torchvision/transforms/v2/_augment.py	191	# after an image or video. However, we need to handle them in _transform, so we make sure to set them to True	COMMENT
MEDIUM	scripts/release_notes/retrieve_prs_data.py	207	# Then you can use classify_prs.py (as a notebook)	STRING

AI Response Leakage1 hit · 8 pts

Severity	File	Line	Snippet	Context
HIGH	references/depth/stereo/utils/losses.py	305	# which should be consistent with the previous step. In this implementation, we simplify the overall loss	COMMENT

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings