Tensors and Dynamic neural networks in Python with strong GPU acceleration
This report presents the forensic synthetic code analysis of pytorch/pytorch, a Python project with 102,650 GitHub stars. SynthScan v2.0 examined 4,309,152 lines of code across 10929 source files, recording 56467 pattern matches distributed across 26 syntactic categories. The overall adjusted score of 12.8 places this repository in the Low AI signal band.
The scanner applied 160+ deterministic lexical heuristics, multi-line block detectors, abstract syntax tree depth profilers, and a cross-file Jaccard similarity matrix to construct a statistically normalised synthetic code estimate. All matches are individually weighted by severity coefficient and contextual multiplier before summation, and the resulting headline score is temporally discounted to account for the repository's development history relative to the commercial emergence of large language model coding tooling (November 2022 onward).
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.
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.
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.
The scanner identified 56467 distinct pattern matches across 26 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.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | .spin/cmds.py | 138 | def regenerate_clangtidy_files(): | CODE |
| LOW | .spin/cmds.py | 246 | def _check_linter_python_versions(): | CODE |
| LOW | .spin/cmds.py | 499 | def regenerate_github_workflows(): | CODE |
| LOW | tools/gen_vulkan_spv.py | 157 | def extract_leading_whitespace(line: str) -> str: | CODE |
| LOW | tools/render_junit.py | 52 | def convert_junit_to_testcases(xml: JUnitXml | TestSuite) -> list[TestCase]: # type: ignore[no-any-unimported] | CODE |
| LOW | tools/create_worktree.py | 108 | def clone_submodule_recursive( | CODE |
| LOW | tools/packaging/build_wheel.py | 53 | def get_supported_python_versions() -> list[str]: | CODE |
| LOW | tools/packaging/build_wheel.py | 74 | def _find_manylinux_interpreters() -> list[str]: | CODE |
| LOW | tools/experimental/torchfuzz/test_determinism.py | 71 | def test_deterministic_output(): | CODE |
| LOW | tools/experimental/torchfuzz/ops_fuzzer.py | 32 | def _get_template_filtered_operators( | CODE |
| LOW | tools/experimental/torchfuzz/visualize_graph.py | 136 | def visualize_operation_graph( | CODE |
| LOW | tools/experimental/torchfuzz/visualize_graph.py | 159 | def operation_graph_to_networkx(graph: OperationGraph): | CODE |
| LOW | tools/experimental/torchfuzz/multi_process_fuzzer.py | 485 | def _print_operation_distribution(results: list[FuzzerResult]) -> None: | CODE |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 190 | def _compute_contiguous_strides(size: tuple[int, ...]) -> list[int]: | CODE |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 213 | def _compute_non_contiguous_dense_strides(size: tuple[int, ...]) -> list[int]: | CODE |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 304 | def _compute_storage_size_needed( | CODE |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 436 | def fuzz_non_contiguous_dense_tensor( | CODE |
| LOW | tools/experimental/torchfuzz/fuzzer.py | 28 | def _parse_supported_ops_with_weights(spec: str) -> tuple[list[str], dict[str, float]]: | CODE |
| LOW | tools/experimental/torchfuzz/fuzzer.py | 57 | def _resolve_generate_only_path( | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 31 | def test_template_inherits_default_ops(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 37 | def test_template_uses_backward_check(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 47 | def test_codegen_creates_streams(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 51 | def test_codegen_has_stream_context(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 55 | def test_codegen_has_final_sync(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 63 | def test_codegen_has_backward(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 67 | def test_codegen_has_requires_grad(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/test_streams_template.py | 72 | def test_codegen_cross_stream_sync(self): | CODE |
| LOW | tools/experimental/torchfuzz/test_streams_template.py | 84 | def test_codegen_event_based_sync(self): | CODE |
| LOW | tools/experimental/torchfuzz/test_streams_template.py | 96 | def test_codegen_deterministic(self): | CODE |
| LOW | tools/experimental/torchfuzz/test_streams_template.py | 102 | def test_codegen_is_valid_python(self): | CODE |
| LOW | tools/experimental/torchfuzz/codegen.py | 345 | def convert_graph_to_python_code( | CODE |
| LOW | tools/experimental/torchfuzz/codegen.py | 517 | def generate_simple_operation_code( | CODE |
| LOW | tools/experimental/torchfuzz/type_promotion.py | 106 | def get_promotion_table_for_strings() -> dict: | CODE |
| LOW | tools/experimental/torchfuzz/type_promotion.py | 162 | def get_scalar_promotion_pairs( | CODE |
| LOW | tools/experimental/torchfuzz/operators/registry.py | 38 | def _concrete_operator_classes(module: ModuleType) -> list[type[Operator]]: | CODE |
| LOW | tools/experimental/torchfuzz/operators/registry.py | 72 | def _register_default_operators(self): | CODE |
| LOW⚡ | tools/experimental/torchfuzz/operators/registry.py | 148 | def set_operator_weight_by_torch_op(torch_op_name: str, weight: float) -> None: | CODE |
| LOW | tools/experimental/torchfuzz/operators/registry.py | 159 | def set_operator_weights_by_torch_op(weights: dict[str, float]) -> None: | CODE |
| LOW | tools/experimental/torchfuzz/cuda/_codegen.py | 567 | def _generate_random_placement(self, tensor_size): | CODE |
| LOW | tools/native_aot/export.py | 323 | def _check_no_orphan_artifacts(out_dir: str, specs=None) -> None: | CODE |
| LOW | tools/native_aot/export.py | 430 | def archs_from_cuda_arch_list(arch_list: str) -> list[str]: | CODE |
| LOW | tools/test/test_test_selections.py | 114 | def test_some_times_with_not_sort_by_time(self) -> None: | CODE |
| LOW | tools/test/test_test_selections.py | 149 | def test_serial_parallel_interleaving(self) -> None: | CODE |
| LOW | tools/test/test_test_selections.py | 435 | def test_disable_pytest_sharding_preserves_file_times(self) -> None: | CODE |
| LOW⚡ | tools/test/test_test_selections.py | 579 | def test_strips_test_osdc_target(self) -> None: | CODE |
| LOW⚡ | tools/test/test_test_selections.py | 587 | def test_build_env_with_test_substring_is_preserved(self) -> None: | CODE |
| LOW⚡ | tools/test/test_test_selections.py | 596 | def test_multi_segment_name_keeps_full_prefix(self) -> None: | CODE |
| LOW⚡ | tools/test/test_test_selections.py | 605 | def test_unrelated_targets_are_not_matched(self) -> None: | CODE |
| LOW | tools/test/test_scoped_library_linter.py | 28 | def test_reports_torch_dot_library_dot_library(self) -> None: | CODE |
| LOW | tools/test/test_scoped_library_linter.py | 53 | def test_reports_imported_library_name(self) -> None: | CODE |
| LOW | tools/test/test_scoped_library_linter.py | 76 | def test_allows_noqa_alone_on_previous_line(self) -> None: | CODE |
| LOW | tools/test/test_scoped_library_linter.py | 88 | def test_allows_multiline_call_if_any_line_has_noqa(self) -> None: | CODE |
| LOW | tools/test/test_scoped_library_linter.py | 102 | def test_reports_multiline_call_without_noqa(self) -> None: | CODE |
| LOW | tools/test/test_scoped_library_linter.py | 118 | def test_empty_on_syntax_error(self) -> None: | CODE |
| LOW | tools/test/test_scoped_library_linter.py | 124 | def test_no_match_for_non_library_call(self) -> None: | CODE |
| LOW | tools/test/test_utils.py | 7 | def test_create_from_namespaced_tuple(self) -> None: | CODE |
| LOW | tools/test/test_utils.py | 18 | def test_namespace_levels_more_than_max(self) -> None: | CODE |
| LOW⚡ | tools/test/test_test_run.py | 24 | def test_union_with_inclusions(self) -> None: | CODE |
| LOW⚡ | tools/test/test_test_run.py | 33 | def test_union_with_non_overlapping_exclusions(self) -> None: | CODE |
| LOW⚡ | tools/test/test_test_run.py | 42 | def test_union_with_overlapping_exclusions(self) -> None: | CODE |
| 32908 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | CMakeLists.txt | 1 | # PyTorch build overview | COMMENT |
| LOW | CMakeLists.txt | 21 | # 4. Dependencies (cmake/Dependencies.cmake): find or build the third-party | COMMENT |
| LOW | CMakeLists.txt | 1041 | # The source code is in utf-8 encoding | COMMENT |
| LOW | pyproject.toml | 1 | # Package ###################################################################### | COMMENT |
| LOW | pyproject.toml | 61 | # Force the Ninja generator on Windows. scikit-build-core otherwise defaults to | COMMENT |
| LOW | pyproject.toml | 81 | # Other than CMAKE_GENERATOR above (win32), cmake.define is intentionally absent | COMMENT |
| LOW | pyproject.toml | 101 | # includes below restore all of them. | COMMENT |
| LOW | pyproject.toml | 161 | COMMENT | |
| LOW | pyproject.toml | 201 | # not consumed at runtime). | COMMENT |
| LOW | pyproject.toml | 261 | # torch/utils/benchmark/examples/. | COMMENT |
| LOW | setup.py | 1 | # PyTorch is built with scikit-build-core through the PEP 517 interface | COMMENT |
| LOW | tools/git_add_generated_dirs.sh | 1 | #!/bin/sh | COMMENT |
| LOW | tools/build_with_debinfo.py | 1 | #!/usr/bin/env python3 | COMMENT |
| LOW | tools/git_reset_generated_dirs.sh | 1 | #!/bin/sh | COMMENT |
| LOW | tools/iwyu/run.sh | 1 | # Run include-what-you-use on a file or folder | COMMENT |
| LOW | tools/native_aot/toolchains.py | 61 | artifact_exts: tuple[str, ...] = () | COMMENT |
| LOW | tools/native_aot/export.py | 61 | os.path.join(os.path.dirname(os.path.abspath(__file__)), "..", "..") | COMMENT |
| LOW | tools/native_aot/export.py | 81 | from torchgen.native_aot_spec_grid import expand_specs | COMMENT |
| LOW | tools/native_aot/export.py | 461 | # Distinct from a declaration's ARCHS (what the KERNELS support, sm_90+); | COMMENT |
| LOW | tools/test/test_native_aot_tools.py | 101 | self.assertEqual(export.export_point.__qualname__, "export_point") | COMMENT |
| LOW | tools/test/test_native_aot_tools.py | 441 | # the tools/*.py glob) and arrive by ordinary import, so only the | COMMENT |
| LOW | tools/test/test_stable_shim_usage_linter.py | 141 | shim_functions = get_shim_functions([sample_shim]) | COMMENT |
| LOW | tools/test/stable_shim_usage_linter_data/sample_usage.h | 141 | // Case 18: Correct, fallback function version is lower than current block, target still equal. | COMMENT |
| LOW | tools/test/stable_shim_usage_linter_data/sample_shim.h | 1 | #ifndef SAMPLE_SHIM_H | COMMENT |
| LOW | tools/test/stable_shim_usage_linter_data/sample_shim.h | 21 | AOTI_TORCH_EXPORT void old_function_2(void* ptr); | COMMENT |
| LOW | tools/test/stable_shim_usage_linter_data/sample_shim.h | 41 | AOTI_TORCH_EXPORT int legacy_fallback(int arg); | COMMENT |
| LOW | tools/test/stable_shim_usage_linter_data/sample_shim.h | 101 | COMMENT | |
| LOW | tools/test/header_only_linter_testdata/umbrella.h | 1 | #pragma once | COMMENT |
| LOW | tools/linter/adapters/flake8_linter.py | 1 | # /// script | COMMENT |
| LOW | tools/linter/adapters/flake8_linter.py | 201 | # "B901": `return x` inside a generator | COMMENT |
| LOW | tools/linter/adapters/raw_throw_linter.py | 41 | # does not cover. An empty prefix means the justification holds everywhere. Add | COMMENT |
| LOW | tools/linter/adapters/pyproject_linter.py | 161 | return [ | COMMENT |
| LOW | tools/linter/adapters/pyproject_linter.py | 181 | # } | COMMENT |
| LOW | tools/linter/adapters/ruff_linter.py | 161 | COMMENT | |
| LOW | tools/linter/adapters/pyrefly_linter.py | 1 | # /// script | COMMENT |
| LOW | tools/linter/adapters/pyrefly_linter.py | 21 | # "optree==0.17.0", | COMMENT |
| LOW | tools/autograd/load_derivatives.py | 341 | "forward definition of gradient as element_wise but it does not " | COMMENT |
| LOW | tools/autograd/load_derivatives.py | 561 | r.name for r in f.func.returns | COMMENT |
| LOW | tools/autograd/gen_python_functions.py | 1 | # Generates Python bindings for ATen functions | COMMENT |
| LOW | tools/autograd/gen_python_functions.py | 21 | # | COMMENT |
| LOW | tools/autograd/gen_python_functions.py | 1181 | return sort_overloads(grouped, symint=symint) | COMMENT |
| LOW | tools/autograd/gen_python_functions.py | 1201 | # If they have different number of parameters then they are not ambiguous - but | COMMENT |
| LOW | tools/autograd/gen_autograd_functions.py | 81 | void will_release_variables() override { | COMMENT |
| LOW | tools/autograd/gen_variable_type.py | 1 | # Generates VariableType.h/cpp | COMMENT |
| LOW | tools/autograd/gen_variable_type.py | 421 | # | COMMENT |
| LOW | tools/autograd/gen_variable_type.py | 1261 | raise AssertionError("info is None in guard_for") | COMMENT |
| LOW | tools/autograd/gen_trace_type.py | 241 | # This replacement implemented in-place with minimum modifications of | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 1 | # Defines derivative formulas and Python signatures of methods on Variable | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 21 | # of how to register separate derivates for different dispatch keys). | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 41 | # - If that argument is differentiable, in the sense that a gradient with respect | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 61 | # are in scope: | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 81 | # | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 101 | # (NB: if your function computes gradient for a list of tensors, | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 121 | # Another interpretation of this expression would read as `var_0` is needed | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 141 | # | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 161 | # (implemented using zeros_like). These gradients are (hopefully) not | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 181 | # according to these three cases: | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 201 | # It is an efficient way to represent a Tensor filled with zeros because | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 1181 | # | list between two `b`s. E.g., | COMMENT |
| LOW | tools/autograd/derivatives.yaml | 1841 | # - name: to_sparse.sparse_dim(Tensor self, int sparse_dim) -> Tensor | COMMENT |
| 5557 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/extract_scripts.py | 3 | CODE | |
| LOW | tools/nvcc_fix_deps.py | 16 | CODE | |
| LOW | tools/gen_vulkan_spv.py | 3 | CODE | |
| LOW | tools/gen_cupti_stubs.py | 24 | CODE | |
| LOW | tools/embed_libomp_macos.py | 16 | CODE | |
| LOW | tools/render_junit.py | 3 | CODE | |
| LOW | tools/strtobool.py | 3 | CODE | |
| LOW | tools/generate_torch_version.py | 1 | CODE | |
| LOW | tools/build_with_debinfo.py | 17 | CODE | |
| LOW | tools/nightly.py | 49 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 4 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 4 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 4 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 4 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 4 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 4 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 12 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 12 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 12 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 13 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 13 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 13 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 14 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 14 | CODE | |
| LOW | tools/experimental/torchfuzz/__init__.py | 14 | CODE | |
| LOW | …ols/experimental/torchfuzz/operators/loss_functions.py | 18 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/bitwise.py | 13 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/logical.py | 8 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/reduction.py | 33 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/activations.py | 33 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 3 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 4 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 5 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 6 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 7 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 8 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 9 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 10 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 20 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 20 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 20 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 20 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 26 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 36 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 36 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 36 | CODE | |
| 5003 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | .spin/cmds.py | 115 | CODE | |
| LOW | .spin/cmds.py | 344 | CODE | |
| LOW | tools/extract_scripts.py | 52 | CODE | |
| LOW | tools/clean.py | 7 | CODE | |
| LOW | tools/gen_vulkan_spv.py | 315 | CODE | |
| LOW | tools/gen_vulkan_spv.py | 376 | CODE | |
| LOW | tools/stale_issues.py | 12 | CODE | |
| LOW | tools/stale_issues.py | 81 | CODE | |
| LOW | tools/gen_cupti_stubs.py | 113 | CODE | |
| LOW | tools/embed_libomp_macos.py | 28 | CODE | |
| LOW | tools/render_junit.py | 62 | CODE | |
| LOW | tools/generate_torch_version.py | 51 | CODE | |
| LOW | tools/download_mnist.py | 40 | CODE | |
| LOW | tools/create_worktree.py | 187 | CODE | |
| LOW | tools/nightly.py | 891 | CODE | |
| LOW | tools/nightly.py | 1160 | CODE | |
| LOW | tools/nightly.py | 320 | CODE | |
| LOW | tools/experimental/torchfuzz/test_determinism.py | 71 | CODE | |
| LOW | tools/experimental/torchfuzz/ops_fuzzer.py | 254 | CODE | |
| LOW | tools/experimental/torchfuzz/visualize_graph.py | 41 | CODE | |
| LOW | tools/experimental/torchfuzz/visualize_graph.py | 159 | CODE | |
| LOW | tools/experimental/torchfuzz/multi_process_fuzzer.py | 83 | CODE | |
| LOW | tools/experimental/torchfuzz/multi_process_fuzzer.py | 217 | CODE | |
| LOW | tools/experimental/torchfuzz/multi_process_fuzzer.py | 517 | CODE | |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 99 | CODE | |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 213 | CODE | |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 320 | CODE | |
| LOW | tools/experimental/torchfuzz/tensor_fuzzer.py | 484 | CODE | |
| LOW | tools/experimental/torchfuzz/fuzzer.py | 80 | CODE | |
| LOW | tools/experimental/torchfuzz/fuzzer.py | 278 | CODE | |
| LOW | tools/experimental/torchfuzz/codegen.py | 517 | CODE | |
| LOW | tools/experimental/torchfuzz/codegen.py | 198 | CODE | |
| LOW | …s/experimental/torchfuzz/operators/tensor_pointwise.py | 61 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/layout.py | 192 | CODE | |
| LOW | tools/experimental/torchfuzz/operators/constant.py | 46 | CODE | |
| LOW | tools/experimental/torchfuzz/cuda/_codegen.py | 159 | CODE | |
| LOW | tools/experimental/torchfuzz/cuda/_codegen.py | 398 | CODE | |
| LOW | tools/experimental/torchfuzz/cuda/_codegen.py | 581 | CODE | |
| LOW | tools/native_aot/export.py | 254 | CODE | |
| LOW | tools/test/test_gen_backend_stubs.py | 27 | CODE | |
| LOW | tools/test/heuristics/test_interface.py | 413 | CODE | |
| LOW | tools/test/heuristics/test_interface.py | 437 | CODE | |
| LOW | tools/linter/adapters/stable_shim_usage_linter.py | 35 | CODE | |
| LOW | tools/linter/adapters/stable_shim_usage_linter.py | 141 | CODE | |
| LOW | tools/linter/adapters/pyfmt_linter.py | 150 | CODE | |
| LOW | tools/linter/adapters/generated_shims_version_linter.py | 134 | CODE | |
| LOW | tools/linter/adapters/no_workflows_on_fork.py | 90 | CODE | |
| LOW | tools/linter/adapters/no_merge_conflict_csv_linter.py | 62 | CODE | |
| LOW | tools/linter/adapters/stable_shim_version_linter.py | 37 | CODE | |
| LOW | tools/linter/adapters/stable_shim_version_linter.py | 97 | CODE | |
| LOW | tools/linter/adapters/stable_shim_version_linter.py | 52 | CODE | |
| LOW | tools/linter/adapters/test_device_bias_linter.py | 45 | CODE | |
| LOW | tools/linter/adapters/test_device_bias_linter.py | 46 | CODE | |
| LOW | tools/linter/adapters/test_device_bias_linter.py | 170 | CODE | |
| LOW | tools/linter/adapters/_stable_shim_utils.py | 334 | CODE | |
| LOW | tools/linter/adapters/raw_throw_linter.py | 111 | CODE | |
| LOW | tools/linter/adapters/raw_throw_linter.py | 239 | CODE | |
| LOW | tools/linter/adapters/raw_throw_linter.py | 324 | CODE | |
| LOW | tools/linter/adapters/raw_throw_linter.py | 404 | CODE | |
| LOW | tools/linter/adapters/header_only_linter.py | 102 | CODE | |
| 3959 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tools/experimental/torchfuzz/codegen.py | 294 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/codegen.py | 296 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 49 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 51 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 99 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 101 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 215 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 217 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 250 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ols/experimental/torchfuzz/operators/loss_functions.py | 252 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 66 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 68 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 185 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 187 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 205 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 207 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 220 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/bitwise.py | 222 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/logical.py | 36 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/logical.py | 38 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/logical.py | 83 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/logical.py | 85 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/logical.py | 103 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/logical.py | 105 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 98 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 100 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 171 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 173 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 210 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 212 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 502 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 504 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 629 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/reduction.py | 631 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/activations.py | 44 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/activations.py | 51 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/activations.py | 109 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/activations.py | 111 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 107 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 109 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 269 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 271 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 317 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 319 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 356 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 358 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 410 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …ls/experimental/torchfuzz/operators/tensor_creation.py | 412 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/comparison.py | 84 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/comparison.py | 86 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/comparison.py | 147 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/comparison.py | 149 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/comparison.py | 207 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/comparison.py | 209 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/experimental/torchfuzz/operators/elementwise_math.py | 33 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/experimental/torchfuzz/operators/elementwise_math.py | 35 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/experimental/torchfuzz/operators/elementwise_math.py | 181 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/experimental/torchfuzz/operators/elementwise_math.py | 183 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/experimental/torchfuzz/operators/elementwise_math.py | 213 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …s/experimental/torchfuzz/operators/elementwise_math.py | 215 | # --------------------------------------------------------------------------- | COMMENT |
| 842 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tools/native_aot/decl.py | 100 | helpers | prelude -> [if (dispatch) { launch; return true; }]* -> return false | STRING |
| HIGH⚡ | tools/test/test_build_with_debinfo.py | 55 | "cd /repo/build/metal && xcrun metal -c /repo/a.metal -o a.air", | CODE |
| HIGH⚡ | tools/test/test_build_with_debinfo.py | 56 | "cd /repo/build/metal && xcrun metallib -o a.metallib a.air", | CODE |
| HIGH | tools/linter/adapters/lintrunner_version_linter.py | 57 | description="Lintrunner is not installed, did you forget to run `make setup-lint && make lint`?", | CODE |
| HIGH⚡ | .ci/lumen_cli/tests/test_run_plan.py | 99 | "export A=x && pytest -q", | CODE |
| HIGH⚡ | .ci/lumen_cli/tests/test_run_plan.py | 100 | "export B=y && pytest -q tests/unit", | CODE |
| HIGH | test/test_datapipe.py | 565 | self.assertTrue(df.equals(cdf)) | CODE |
| HIGH⚡ | test/test_schema_check.py | 466 | # Empty containers return false | COMMENT |
| HIGH⚡ | test/test_jit.py | 6528 | # 1. nan == nan should return true | STRING |
| HIGH⚡ | test/test_pytree.py | 1287 | '"children_spec": [{"type": null, "context": null, ' | STRING |
| HIGH⚡ | test/test_pytree.py | 1288 | '"children_spec": []}, {"type": null, "context": null, ' | STRING |
| HIGH⚡ | test/test_pytree.py | 1289 | '"children_spec": []}]}, {"type": null, "context": null, ' | STRING |
| HIGH⚡ | test/test_pytree.py | 1291 | '"[4, 5, 6]", "children_spec": [{"type": null, "context": null, ' | STRING |
| HIGH⚡ | test/test_pytree.py | 1292 | '"children_spec": []}, {"type": null, "context": null, "children_spec": ' | STRING |
| HIGH⚡ | test/test_pytree.py | 1293 | '[]}, {"type": null, "context": null, "children_spec": []}]}]}]' | STRING |
| HIGH | test/test_cuda.py | 11006 | if (row < M && col < N) { | CODE |
| HIGH | test/test_opaque_obj_v2.py | 477 | queue.push(b) | CODE |
| HIGH | test/test_meta.py | 1342 | with MetaCrossRefDispatchMode.push( | STRING |
| HIGH⚡ | test/inductor/test_torchbind.py | 122 | q.push(torch.ones(2, 3)) | CODE |
| HIGH | test/inductor/test_torchbind.py | 350 | self.tq.push(x.cos()) | CODE |
| HIGH | test/inductor/test_torchbind.py | 351 | self.tq.push(x.sin()) | CODE |
| HIGH | test/cpp_extensions/test_libtorch_agnostic.py | 2327 | # Runtime predates the shim: dlsym returns null, so we fall back to | COMMENT |
| HIGH | test/nn/test_pooling.py | 1060 | "MaxUnpool2d": r"Assertion `maxind >= 0 && maxind < outputImageSize` failed", | CODE |
| HIGH | test/nn/test_pooling.py | 1061 | "MaxUnpool3d": r"Assertion `index >= 0 && index < outputImageSize` failed", | CODE |
| HIGH | test/nn/test_embedding.py | 1191 | msg = "Expected idx >= 0 && idx < num_embeddings" | CODE |
| HIGH | test/nn/test_embedding.py | 1246 | "embedding_idx >= 0 && embedding_idx < num_embeddings" | STRING |
| HIGH | test/nn/test_embedding.py | 1248 | else "bag_idx >= 0 && bag_idx < num_bags" | STRING |
| HIGH | test/distributed/test_distributed_spawn.py | 62 | f"e.g.\ntouch /tmp/barrier && TEMP_DIR=/tmp BACKEND='nccl' WORLD_SIZE=2 python {__file__}", | CODE |
| HIGH⚡ | test/distributed/test_fake_pg.py | 76 | # default init path leaves the backend options null). | COMMENT |
| HIGH | test/fx/test_fx_split.py | 211 | # pyre-fixme[61]: `name` is undefined, or not always defined. | COMMENT |
| HIGH | test/quantization/eager/test_equalize_eager.py | 56 | identical modules should return true | STRING |
| HIGH | test/quantization/eager/test_equalize_eager.py | 57 | modules with high difference in weights should return false | STRING |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 812 | stack.push(_expect_ok) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 815 | stack.push(cm) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 817 | stack.push(_suppress_exc) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 820 | stack.push(cm) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 822 | stack.push(_expect_exc) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 824 | stack.push(_expect_exc) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 893 | stack.push(lambda *exc: False) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 898 | stack.push(lambda *exc: True) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 1066 | stack.push(suppress_exc) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 1137 | stack.push(suppress_exc) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 1144 | stack.push(lambda *exc: True) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 1145 | stack.push(lambda *exc: 1/0) | CODE |
| HIGH⚡ | test/cpython/v3_13/test_contextlib.py | 1146 | stack.push(lambda *exc: {}[1]) | CODE |
| HIGH | test/cpython/v3_13/test_contextlib.py | 997 | stack.push(suppress_exc) | CODE |
| HIGH | test/cpython/v3_13/test_contextlib.py | 1163 | stack.push(cm) | CODE |
| HIGH⚡ | test/jit/test_torchbind.py | 98 | stackstring.push("lel") | CODE |
| HIGH | test/jit/test_torchbind.py | 61 | ss1.push(ss2.pop()) | CODE |
| HIGH | test/lazy/test_functionalization.py | 98 | %1 = [BFloat16[3]] aten::_to_copy(%0), dtype=BFloat16, layout=null, device=null, pin_memory=null, non_blocking=0, memo | STRING |
| HIGH⚡ | test/export/test_torchbind.py | 1110 | tq.push(torch.zeros(2, 2)) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1111 | tq.push(torch.ones(2, 2)) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1113 | tq.push(torch.ones(2, 2) * 2) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1181 | tq.push(x.cos()) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1182 | tq.push(x.sin()) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1242 | tq.push(x.cos()) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1243 | tq.push(x.sin()) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1257 | tq2.push(torch.randn(4, 5, requires_grad=False)) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1263 | tq3.push(torch.randn(2, 3, requires_grad=False)) | CODE |
| HIGH⚡ | test/export/test_torchbind.py | 1269 | tq4.push(torch.randn(2, 3, requires_grad=False)) | CODE |
| 439 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tools/update_masked_docs.py | 30 | # This file is generated, do not modify it! | STRING |
| MEDIUM | tools/experimental/torchfuzz/ops_fuzzer.py | 472 | # Create the operation node | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/multi_process_fuzzer.py | 22 | # Create a mock tqdm class for type safety | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/tensor_fuzzer.py | 251 | # Create a valid permutation that's still dense | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/tensor_fuzzer.py | 344 | # Create a local Random instance to avoid interfering with global state | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/tensor_fuzzer.py | 499 | # Create a local random instance to avoid interfering with global state | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/tensor_fuzzer.py | 507 | # Create a scalar value based on dtype | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/tensor_fuzzer.py | 527 | # Create a scalar value based on dtype | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/fuzzer.py | 129 | # Create a unique folder for this iteration | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/operators/item.py | 27 | # Create a tensor spec that can produce a scalar via .item() | COMMENT |
| MEDIUM | tools/test/test_gb_registry_linter.py | 438 | # Create a callsite with one of the duplicate types | COMMENT |
| MEDIUM⚡ | tools/test/heuristics/test_interface.py | 29 | # Create a dummy heuristic class | COMMENT |
| MEDIUM | tools/linter/adapters/stable_shim_usage_linter.py | 128 | "# This file is automatically updated by the stable_shim_usage_linter.\n" | CODE |
| MEDIUM | tools/linter/adapters/s3_init.py | 122 | # Create the output folder | COMMENT |
| MEDIUM | tools/autograd/gen_variable_type.py | 432 | # The following code templates implement the checks for this invariant: | COMMENT |
| MEDIUM | tools/autograd/derivatives.yaml | 1499 | # Defining a backward will make codegen spit out the forward call as | COMMENT |
| MEDIUM | tools/setup_helpers/gen_version_header.py | 59 | # Create the output dir if it doesn't exist. | COMMENT |
| MEDIUM | .ci/lumen_cli/tests/test_utils.py | 26 | # Create a temporary directory for the test to run in | COMMENT |
| MEDIUM | .ci/lumen_cli/cli/lib/core/vllm/vllm_test_library.yaml | 2 | # This file defines the test plans for vllm CI development | COMMENT |
| MEDIUM | .ci/libtorch/extract_libtorch_from_wheel.py | 520 | # Create the zip | COMMENT |
| MEDIUM | test/test_python_dispatch.py | 743 | # Create a fragment | STRING |
| MEDIUM | test/test_linalg.py | 10360 | # Create the results file in the parent so we can always clean it up, | STRING |
| MEDIUM | test/test_functionalization.py | 1825 | # Create a view of x | COMMENT |
| MEDIUM⚡ | test/test_hub.py | 316 | # Create an entry with malicious path | COMMENT |
| MEDIUM⚡ | test/test_hub.py | 322 | # Create a single entry with malicious path for legacy format | COMMENT |
| MEDIUM⚡ | test/test_hub.py | 390 | # Create a legacy-style malicious zip (single entry with traversal) | COMMENT |
| MEDIUM | test/test_hub.py | 412 | # Create a zip that looks like a GitHub repository but has malicious content | COMMENT |
| MEDIUM | test/test_matmul_cuda.py | 1072 | # Create a doubly-non-contiguous 2D matrix: stride on both dims > 1 | COMMENT |
| MEDIUM | test/test_torch.py | 3528 | # Create the 4 possible combinations of scalar sizes for source / index | COMMENT |
| MEDIUM | test/test_torch.py | 3577 | # Create the 8 possible combinations of scalar sizes for target / index / source | COMMENT |
| MEDIUM | test/test_torch.py | 8856 | # This method is primarily exposed for torchvision's resize | COMMENT |
| MEDIUM | test/test_torch.py | 11158 | # The following block extends TestTorch with negative dim wrapping tests | COMMENT |
| MEDIUM⚡ | test/test_dynamic_shapes.py | 6916 | # Create a backed symbol | COMMENT |
| MEDIUM⚡ | test/test_dynamic_shapes.py | 6921 | # Create an unbacked symbol | COMMENT |
| MEDIUM | test/test_dynamic_shapes.py | 4309 | # Create a symint during fake tensor tracing | COMMENT |
| MEDIUM | test/test_dynamic_shapes.py | 5416 | # Create a scalar tensor to scatter into the selected index | COMMENT |
| MEDIUM | test/test_dynamic_shapes.py | 6444 | # Create a tensor with shape [..., 2] where batch dim is unbacked | COMMENT |
| MEDIUM | test/test_dynamic_shapes.py | 6862 | # Create a derived expression: u0 // 2 | COMMENT |
| MEDIUM | test/test_dynamic_shapes.py | 7006 | # Create a backed foreign symbol that this env cannot translate. | COMMENT |
| MEDIUM | test/test_optim.py | 2160 | # Create a random 2D tensor for compatibility with Muon. | COMMENT |
| MEDIUM | test/test_optim.py | 2219 | # Create an arbitrary 2D tensor for compatibility with Muon. | COMMENT |
| MEDIUM⚡ | test/test_nn.py | 8068 | # Create an appropriately-sized input with a single spatial element. | COMMENT |
| MEDIUM | test/test_subclass.py | 231 | # Define a subclass that does not rewrap for any function in its __torch_dispatch__ impl. | COMMENT |
| MEDIUM | test/test_indexing.py | 1981 | # Create the 8 possible combinations of scalar sizes for target / index / source | COMMENT |
| MEDIUM | test/test_indexing.py | 2327 | # Create the 4 possible combinations of scalar sizes for index / source | COMMENT |
| MEDIUM | test/test_custom_ops.py | 1542 | # Define a named tuple for a Point with x and y coordinates | COMMENT |
| MEDIUM | test/test_cuda.py | 3043 | # Define a function to create generator states, with optional graph registration | COMMENT |
| MEDIUM | test/test_cuda.py | 3057 | # Define a function to perform specific RNG actions using the generator's states | COMMENT |
| MEDIUM | test/test_cuda.py | 3074 | # Define a function to retrieve the final offsets of the original and new generator states | COMMENT |
| MEDIUM | test/test_cuda.py | 5734 | # Create a pipe to communicate between parent and child processes | COMMENT |
| MEDIUM | test/test_cuda.py | 6168 | # Create a simple model with a backward pass | COMMENT |
| MEDIUM | test/test_cuda.py | 6358 | # Create an instance of the model | COMMENT |
| MEDIUM | test/test_cuda.py | 6363 | # Create a sample input tensor | COMMENT |
| MEDIUM | test/test_cuda.py | 2588 | # This function is intended to test the lazy creation and reuse of per-thread | STRING |
| MEDIUM | test/test_cuda.py | 11072 | # Define a simple vector addition kernel | STRING |
| MEDIUM | test/test_opaque_obj_v2.py | 106 | # Create a class with an unsupported descriptor | COMMENT |
| MEDIUM | test/test_opaque_obj_v2.py | 225 | # Create a new counter to match device mesh's __getitem__ | COMMENT |
| MEDIUM⚡ | test/test_dlpack.py | 445 | # Create a new tensor, and wrap it using TensorDLPackWrapper. | COMMENT |
| MEDIUM⚡ | test/test_dlpack.py | 448 | # Create a new tensor from the wrapper. | COMMENT |
| MEDIUM | test/test_dlpack.py | 123 | # Create a stream where the tensor will reside | COMMENT |
| 611 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/render_junit.py | 33 | except Exception as err: | CODE |
| MEDIUM | tools/nightly_hotpatch.py | 94 | print("Error: PyTorch is not installed in the current Python environment.") | CODE |
| LOW | tools/nightly_hotpatch.py | 130 | except Exception as e: | CODE |
| MEDIUM | tools/nightly_hotpatch.py | 131 | print(f"An error occurred while downloading the patch: {e}") | CODE |
| MEDIUM⚡ | tools/nightly_hotpatch.py | 178 | print("Error: The 'patch' utility is not installed or not found in PATH.") | CODE |
| LOW⚡ | tools/nightly_hotpatch.py | 180 | except Exception as e: | CODE |
| MEDIUM⚡ | tools/nightly_hotpatch.py | 181 | print(f"An error occurred while applying the patch: {e}") | CODE |
| LOW | tools/generate_torch_version.py | 30 | except Exception: | CODE |
| LOW | tools/generate_torch_version.py | 47 | except Exception: | CODE |
| LOW | tools/nightly.py | 721 | except Exception as e: | CODE |
| LOW | tools/nightly.py | 945 | except Exception: | CODE |
| LOW | tools/experimental/torchfuzz/ops_fuzzer.py | 238 | except Exception: | CODE |
| LOW | tools/experimental/torchfuzz/multi_process_fuzzer.py | 183 | except Exception as e: | CODE |
| LOW | tools/experimental/torchfuzz/multi_process_fuzzer.py | 363 | except Exception as e: | CODE |
| LOW | tools/experimental/torchfuzz/fuzzer.py | 266 | except Exception as e: | CODE |
| MEDIUM | tools/experimental/torchfuzz/fuzzer.py | 274 | print(f"Error: {error_message}") | CODE |
| LOW | tools/experimental/torchfuzz/fuzzer.py | 437 | except Exception as e: | CODE |
| LOW | tools/experimental/torchfuzz/fuzzer.py | 476 | except Exception as e: | CODE |
| LOW | tools/test/gen_operators_yaml_test.py | 141 | except Exception: | CODE |
| LOW | tools/linter/adapters/gha_linter.py | 61 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/actionlint_linter.py | 168 | except Exception: | STRING |
| LOW | tools/linter/adapters/grep_linter.py | 136 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/grep_linter.py | 216 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/grep_linter.py | 397 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/cmake_linter.py | 144 | except Exception: | STRING |
| LOW | tools/linter/adapters/pyfmt_linter.py | 138 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/pyfmt_linter.py | 193 | except Exception: | CODE |
| LOW | tools/linter/adapters/no_workflows_on_fork.py | 242 | except Exception: | CODE |
| LOW | tools/linter/adapters/no_merge_conflict_csv_linter.py | 97 | except Exception: | CODE |
| LOW | tools/linter/adapters/stable_shim_version_linter.py | 89 | except Exception as e: | CODE |
| LOW | tools/linter/adapters/nativefunctions_linter.py | 75 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/clangformat_linter.py | 260 | except Exception: | CODE |
| LOW | tools/linter/adapters/docstring_linter.py | 92 | except Exception as e: | STRING |
| MEDIUM | tools/linter/adapters/docstring_linter.py | 86 | def _grandfather(self) -> dict[str, dict[str, Any]]: | CODE |
| LOW | tools/linter/adapters/pyproject_linter.py | 250 | except Exception: | CODE |
| LOW | tools/linter/adapters/clangtidy_linter.py | 335 | except Exception: | STRING |
| LOW | tools/linter/adapters/cmake_minimum_required_linter.py | 184 | except Exception: | CODE |
| LOW | tools/linter/adapters/cmake_minimum_required_linter.py | 244 | except Exception: | CODE |
| LOW | tools/linter/adapters/ruff_linter.py | 479 | except Exception: # Catch all exceptions for lintrunner | CODE |
| LOW | tools/linter/adapters/newlines_linter.py | 96 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/newlines_linter.py | 135 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/codespell_linter.py | 140 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/codespell_linter.py | 167 | except Exception as err: | CODE |
| LOW | tools/linter/adapters/codespell_linter.py | 224 | except Exception: | CODE |
| LOW | tools/dynamo/verify_dynamo.py | 187 | except Exception: | CODE |
| LOW | tools/testing/upload_artifacts.py | 121 | except Exception as e: | CODE |
| LOW⚡ | tools/testing/upload_artifacts.py | 207 | except Exception as e: | CODE |
| LOW⚡ | tools/testing/upload_artifacts.py | 219 | except Exception as e: | CODE |
| LOW | tools/testing/target_determination/determinator.py | 30 | except Exception as e: | CODE |
| MEDIUM | tools/testing/target_determination/determinator.py | 31 | print(f"Error in {heuristic.__class__.__name__}: {e}", file=file) | CODE |
| LOW⚡ | …ing/target_determination/heuristics/mentioned_in_pr.py | 33 | except Exception as e: | CODE |
| LOW⚡ | …ing/target_determination/heuristics/mentioned_in_pr.py | 42 | except Exception as e: | CODE |
| LOW | …ing/target_determination/heuristics/mentioned_in_pr.py | 53 | except Exception: | CODE |
| LOW | tools/torchtlx/sanity.py | 29 | except Exception as e: # report every check, do not abort on the first | CODE |
| LOW | tools/torchtlx/_probe.py | 59 | except Exception as e: | CODE |
| LOW | tools/torchtlx/_probe.py | 73 | except Exception as e: | CODE |
| LOW | tools/torchtlx/_probe.py | 88 | except Exception as e: | CODE |
| LOW | tools/stats/monitor.py | 244 | except Exception as e: | CODE |
| LOW | tools/stats/monitor.py | 323 | except Exception as e: | CODE |
| LOW | tools/stats/monitor.py | 472 | except Exception as e: | CODE |
| 1364 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL | tools/autograd/load_derivatives.py | 292 | is_foreach = f.func.name.name.base.startswith("_foreach_") | CODE |
| CRITICAL | tools/autograd/gen_autograd_functions.py | 694 | info.func.func.name.name.base.startswith("_foreach") and is_output | STRING |
| CRITICAL | test/test_per_overload_api.py | 73 | torch.ops.aten.linear.default.decompose(x, y), | CODE |
| CRITICAL | test/inductor/test_gpu_select_algorithm.py | 104 | w_int8 = torch.ao.quantization.fx._decomposed.quantize_per_channel( | CODE |
| CRITICAL | test/inductor/test_gpu_select_algorithm.py | 159 | w_int8 = torch.ao.quantization.fx._decomposed.quantize_per_channel( | CODE |
| CRITICAL | test/inductor/test_cpu_select_algorithm.py | 1734 | w_int8 = torch.ao.quantization.fx._decomposed.quantize_per_channel( | CODE |
| CRITICAL | test/inductor/test_cpu_select_algorithm.py | 1802 | w_int8 = torch.ao.quantization.fx._decomposed.quantize_per_channel( | CODE |
| CRITICAL | test/onnx/test_pytorch_onnx_onnxruntime.py | 12902 | model = torch.ao.nn.intrinsic.quantized.ConvReLU1d(16, 33, 3, stride=2) | CODE |
| CRITICAL | test/onnx/test_pytorch_onnx_onnxruntime.py | 12916 | model = torch.ao.nn.intrinsic.quantized.ConvReLU2d(16, 33, 3, stride=2) | CODE |
| CRITICAL | test/onnx/test_pytorch_onnx_onnxruntime.py | 12931 | model = torch.ao.nn.intrinsic.quantized.ConvReLU3d( | CODE |
| CRITICAL | test/fx/test_dce_pass.py | 235 | torch._ops.ops.aten.add_.Tensor(a, 1) | CODE |
| CRITICAL | test/fx/test_dce_pass.py | 317 | torch._ops.ops.aten.add.out(b, b, out=a, alpha=2) | CODE |
| CRITICAL | test/dynamo/test_minifier.py | 346 | conv = torch.ao.nn.intrinsic.qat.ConvBnReLU2d( | CODE |
| CRITICAL | test/dynamo/test_minifier.py | 368 | "torch.ao.nn.intrinsic.qat.modules.conv_fused.ConvBnReLU2d(", | CODE |
| CRITICAL | test/dynamo/test_minifier.py | 396 | conv = torch.ao.nn.intrinsic.qat.ConvBnReLU2d( | CODE |
| CRITICAL | test/dynamo/test_reconstruct.py | 376 | return self.a.b.c.d.e(x) | CODE |
| CRITICAL | test/quantization/core/test_quantized_module.py | 1742 | cell_dq = torch.ao.nn.quantized.dynamic.LSTM(input_size=input_size, | STRING |
| CRITICAL | test/quantization/core/test_quantized_module.py | 1750 | ref_dq = torch.ao.nn.quantized.dynamic.LSTM(input_size=input_size, | STRING |
| CRITICAL | test/quantization/core/test_quantized_module.py | 1804 | cell_dq = torch.ao.nn.quantized.dynamic.GRU(input_size=input_size, | STRING |
| CRITICAL | test/quantization/core/test_quantized_op.py | 678 | qY = torch.ao.nn.quantized.functional.elu(qX, output_scale, output_zero_point, alpha=alpha) | CODE |
| CRITICAL | test/quantization/core/test_quantized_op.py | 1037 | qY = torch.ao.nn.quantized.functional.hardswish( | CODE |
| CRITICAL | test/quantization/core/test_quantized_op.py | 3055 | qY = torch.ao.nn.quantized.functional.hardtanh(qX, -1, 6) | CODE |
| CRITICAL | test/quantization/core/test_quantized_op.py | 9497 | qY_hat = torch.ao.nn.quantized.functional.hardtanh(qX, min_val, max_val) | CODE |
| CRITICAL | test/quantization/fx/test_quantize_fx.py | 4944 | return torch.ao.quantization.fx.lstm_utils._get_lstm_with_individually_observed_parts( | CODE |
| CRITICAL | test/quantization/fx/test_quantize_fx.py | 4964 | return torch.ao.quantization.fx.lstm_utils._get_reference_quantized_lstm_module( | CODE |
| CRITICAL | test/quantization/fx/test_quantize_fx.py | 6770 | activation=torch.ao.quantization.observer.HistogramObserver.with_args( | CODE |
| CRITICAL⚡ | test/quantization/jit/test_quantize_jit.py | 3844 | eager_model.sub.module.fc1.weight.detach() | CODE |
| CRITICAL⚡ | test/quantization/jit/test_quantize_jit.py | 3847 | eager_model.sub.module.fc1.bias.detach() | CODE |
| CRITICAL⚡ | test/quantization/jit/test_quantize_jit.py | 3850 | eager_model.sub.module.fc2.weight.detach() | CODE |
| CRITICAL⚡ | test/quantization/jit/test_quantize_jit.py | 3853 | eager_model.sub.module.fc2.bias.detach() | CODE |
| CRITICAL | test/quantization/jit/test_quantize_jit.py | 3798 | eager_model.sub2.fc1.module.weight.detach() | STRING |
| CRITICAL | test/quantization/jit/test_quantize_jit.py | 3801 | eager_model.sub2.fc1.module.bias.detach() | STRING |
| CRITICAL | test/quantization/eager/test_quantize_eager_ptq.py | 1078 | original_m.sub.custom.conv.weight.detach() | STRING |
| CRITICAL | test/quantization/eager/test_quantize_eager_ptq.py | 1081 | original_m.sub.custom.conv.bias.detach() | STRING |
| CRITICAL | test/cpp/jit/test_backend.cpp | 49 | __torch__.torch.classes.__backends__.test_backend.__new__(__torch__.torch.classes.__backends__.test_backend) | CODE |
| CRITICAL | test/jit/test_class_type.py | 556 | three = jit._imported_class_test.very.very.nested.FooUniqueName(a) | CODE |
| CRITICAL | test/export/test_torchbind.py | 348 | torch.ops._TorchScriptTesting.takes_foo.default._dispatch_cache.clear() | CODE |
| CRITICAL | test/export/test_torchbind.py | 359 | torch.ops._TorchScriptTesting.takes_foo.default.py_impl( | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13079 | x = self.n3.n4.n5.n6.n7.n8(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13081 | self.n3.n4.n5.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13082 | self.n3.n4.n5.n6.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13083 | self.n3.n4.n5.n6.n7.n8.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13118 | x = self.n1.n2.n3.n4.n5.n6.n7(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13119 | self.n1.n2.n3.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13120 | self.n1.n2.n3.n4.n5.n6.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13121 | self.n1.n2.n3.n4.n5.n6.n7.n8.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13498 | x = self.n1.n2.n3.n4.n5(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13499 | x = self.n1.n2.n3.n4.n5.n6(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13501 | self.n1.n2.n3.n4.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13502 | self.n1.n2.n3.n4.n5.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13503 | self.n1.n2.n3.n4.n5.n6.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13597 | x = self.n5.n6.n7.n8.n9(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13600 | self.n5.n6.n7.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13601 | self.n5.n6.n7.n8.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13618 | x = self.n4.n5.n6.n7.n8(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13619 | x = self.n4.n5.n6.n7.n8.n9(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13620 | self.n4.n5.n6.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13621 | self.n4.n5.n6.n7.n8.buf.add_(1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13638 | x = self.n3.n4.n5.n6.n7.n8(x + 1) | CODE |
| CRITICAL⚡ | test/export/test_export.py | 13639 | x = self.n3.n4.n5.n6.n7.n8.n9(x + 1) | CODE |
| 75 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/build_libtorch.py | 17 | def _set_cmake_build_type() -> None: | CODE |
| LOW | tools/packaging/build_wheel.py | 19 | logger = logging.getLogger(__name__) | CODE |
| LOW | tools/experimental/torchfuzz/__init__.py | 17 | __all__ = [ | CODE |
| LOW | tools/experimental/torchfuzz/operators/registry.py | 116 | def set_operator_weight(op_name: str, weight: float) -> None: | CODE |
| LOW⚡ | tools/experimental/torchfuzz/operators/registry.py | 142 | def set_operator_weights(weights: dict[str, float]) -> None: | CODE |
| LOW⚡ | tools/experimental/torchfuzz/operators/registry.py | 148 | def set_operator_weight_by_torch_op(torch_op_name: str, weight: float) -> None: | CODE |
| LOW | tools/experimental/torchfuzz/operators/registry.py | 159 | def set_operator_weights_by_torch_op(weights: dict[str, float]) -> None: | CODE |
| LOW | tools/experimental/torchfuzz/operators/__init__.py | 62 | __all__ = [ | CODE |
| LOW | tools/native_aot/decl.py | 119 | __all__ = [ | CODE |
| LOW | tools/linter/adapters/_linter/__init__.py | 13 | __all__ = [ | CODE |
| LOW | tools/linter/clang_tidy/generate_build_files.py | 25 | def update_submodules() -> None: | CODE |
| LOW | …s/testing/target_determination/heuristics/interface.py | 78 | def set_test_score(self, test_run: TestRun, new_score: float) -> None: | CODE |
| LOW | tools/code_coverage/package/tool/summarize_jsons.py | 189 | def update_set() -> None: | CODE |
| LOW | tools/metadata/version.py | 22 | __all__ = ["dynamic_metadata"] | CODE |
| LOW | tools/metadata/dependencies.py | 25 | __all__ = ["dynamic_metadata", "dynamic_wheel"] | CODE |
| LOW | .ci/lumen_cli/cli/run.py | 11 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/test_cli/register_test.py | 11 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/core/vllm/lib.py | 24 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/core/vllm/vllm_test.py | 25 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/core/vllm/vllm_build.py | 33 | logger = logging.getLogger(__name__) | CODE |
| LOW⚡ | .ci/lumen_cli/cli/lib/core/torchtitan/lib.py | 11 | logger = logging.getLogger(__name__) | CODE |
| LOW | …i/lumen_cli/cli/lib/core/torchtitan/torchtitan_test.py | 16 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/common/docker_helper.py | 13 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/common/gh_summary.py | 17 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/common/utils.py | 16 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/common/path_helper.py | 10 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/common/pip_helper.py | 14 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/lib/common/git_helper.py | 12 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/lumen_cli/cli/build_cli/register_build.py | 10 | logger = logging.getLogger(__name__) | CODE |
| LOW | .ci/manywheel/repair_wheel.py | 304 | def set_rpath(sofile: Path, rpath: str, force_rpath: bool) -> None: | CODE |
| LOW | test/test_fx_passes.py | 21 | logger = logging.getLogger(__name__) | CODE |
| LOW | test/inductor/test_aoti_cross_compile_windows.py | 51 | def set_base_path(cls, path: Path | str | None = None) -> None: | CODE |
| LOW | test/inductor/test_aoti_cross_compile_windows.py | 56 | def set_win_torch_libs_path(cls, path: str | None = None) -> None: | CODE |
| LOW | …inductor/extension_backends/triton/device_interface.py | 57 | def set_device(device: int) -> None: | CODE |
| LOW | …inductor/extension_backends/triton/device_interface.py | 77 | def set_device(device) -> None: | CODE |
| LOW | …ension/torch_openreg/torch_openreg/openreg/__init__.py | 45 | def set_device(device) -> None: | STRING |
| LOW | …ension/torch_openreg/torch_openreg/openreg/__init__.py | 77 | __all__ = [ | STRING |
| LOW | …xtension/torch_openreg/torch_openreg/openreg/random.py | 7 | __all__ = [ | CODE |
| LOW | …torch_agn_2_13_extension/libtorch_agn_2_13/__init__.py | 26 | __all__ = [ | CODE |
| LOW | …torch_agn_2_14_extension/libtorch_agn_2_14/__init__.py | 22 | __all__ = [ | CODE |
| LOW | …ibtorch_agn_2_9_extension/libtorch_agn_2_9/__init__.py | 18 | __all__ = [ | CODE |
| LOW | …torch_agn_2_12_extension/libtorch_agn_2_12/__init__.py | 18 | __all__ = [ | CODE |
| LOW | …torch_agn_2_11_extension/libtorch_agn_2_11/__init__.py | 18 | __all__ = [ | CODE |
| LOW | …torch_agn_2_10_extension/libtorch_agn_2_10/__init__.py | 18 | __all__ = [ | CODE |
| LOW | test/distributed/test_c10d_functional_native.py | 722 | def _set_group_name(self, name: str) -> None: | CODE |
| LOW | test/distributed/checkpoint/test_pg_transport.py | 42 | logger = logging.getLogger(__name__) | CODE |
| LOW | test/distributed/checkpoint/test_dtensor_checkpoint.py | 66 | def set_extra_state(self, state: dict[str, int | torch._tensor.Tensor]) -> None: | CODE |
| LOW | test/distributed/checkpoint/test_dtensor_resharding.py | 39 | logger = logging.getLogger(__name__) | CODE |
| LOW | test/distributed/checkpoint/test_checkpoint.py | 215 | def set_up_storage_reader(self, metadata: Metadata, is_coordinator: bool) -> None: | CODE |
| LOW | test/distributed/elastic/test_control_plane.py | 229 | def set_content(self, content: str, content_type: str) -> None: | CODE |
| LOW | test/distributed/elastic/test_control_plane.py | 233 | def set_status(self, status: int) -> None: | CODE |
| LOW | …tributed/elastic/rendezvous/dynamic_rendezvous_test.py | 215 | def set_state_internal(self, state: _RendezvousState) -> None: | CODE |
| LOW | test/distributed/elastic/rendezvous/api_test.py | 216 | def set_closed(self) -> None: | CODE |
| LOW | test/distributed/elastic/utils/util_test.py | 26 | def set_timeout(self, timeout: float) -> None: | CODE |
| LOW | test/distributed/pipelining/test_schedule.py | 75 | logger = logging.getLogger(__name__) | CODE |
| LOW | test/distributed/pipelining/test_schedule_multiproc.py | 66 | logger = logging.getLogger(__name__) | CODE |
| LOW | …/distributed/_composable/fsdp/test_fully_shard_comm.py | 984 | def set_forward_prefetch(model: Transformer, num_to_prefetch: int) -> None: | CODE |
| LOW | …/distributed/_composable/fsdp/test_fully_shard_comm.py | 1079 | def set_backward_prefetch(model: Transformer, num_to_prefetch: int) -> None: | CODE |
| LOW | …/distributed/_composable/fsdp/test_fully_shard_comm.py | 1205 | def set_backward_prefetch(model: Transformer) -> None: | CODE |
| LOW | …t/distributed/tensor/test_dtensor_dispatch_overhead.py | 22 | logger = logging.getLogger(__name__) | CODE |
| 1136 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | test/test_transformers.py | 130 | Compare test tensor against golden and reference tensors. Golden is the highest precision possible serving as t | STRING |
| HIGH | test/onnx/onnx_test_common.py | 172 | Run ORT on the given ONNX model and inputs Used in test_fx_to_onnx_with_onnxruntime.py Args: onnx_mode | STRING |
| HIGH | test/onnx/torchlib/ops_test_common.py | 327 | Run a model with ONNX Runtime in a separate process. Args: serialized_model: Serialized ONNX model proto. | STRING |
| HIGH | .claude/skills/docstring/SKILL.md | 295 | Sample from the Gumbel-Softmax distribution and optionally discretize. Args: logits (Tensor): `[..., n | STRING |
| HIGH | torch/library.py | 1708 | Returns the computed kernel for a given operator and dispatch key. This function retrieves the kernel that would be | STRING |
| HIGH | torch/_torch_docs.py | 4782 | unflatten(input, dim, sizes) -> Tensor Expands a dimension of the input tensor over multiple dimensions. .. seealso:: | STRING |
| HIGH | torch/_torch_docs.py | 6761 | lu_unpack(LU_data, LU_pivots, unpack_data=True, unpack_pivots=True, *, out=None) -> (Tensor, Tensor, Tensor) Unpacks t | STRING |
| HIGH | torch/_torch_docs.py | 7732 | aminmax(input, *, dim=None, keepdim=False, out=None) -> (Tensor min, Tensor max) Computes the minimum and maximum valu | STRING |
| HIGH | torch/_torch_docs.py | 11994 | triangular_solve(b, A, upper=True, transpose=False, unitriangular=False, *, out=None) -> (Tensor, Tensor) Solves a sys | STRING |
| HIGH | torch/functional.py | 80 | broadcast_shapes(*shapes) -> Size Similar to :func:`broadcast_tensors` but for shapes. This is equivalent to | STRING |
| HIGH | torch/functional.py | 176 | einsum(equation, *operands) -> Tensor Sums the product of the elements of the input :attr:`operands` along dimensio | STRING |
| HIGH | torch/_lobpcg.py | 359 | Find the k largest (or smallest) eigenvalues and the corresponding eigenvectors of a symmetric positive definite gen | STRING |
| HIGH | torch/serialization.py | 714 | Restores `storage` using a deserializer function registered for the `location`. This function looks in the reg | STRING |
| HIGH | torch/_functorch/einops.py | 16 | A native implementation of `einops.rearrange`, a reader-friendly smart element reordering for multidimensional tenso | STRING |
| HIGH | torch/_functorch/_aot_autograd/fx_utils.py | 47 | Given a joint graph with descriptors (meta['desc'] on placeholders and output), returns the node for every inpu | STRING |
| HIGH | torch/_functorch/_aot_autograd/fx_utils.py | 103 | Get all output nodes and their corresponding tangent nodes from a joint graph. Similar to get_all_input_and_grad_no | STRING |
| HIGH | torch/_functorch/_aot_autograd/fx_utils.py | 231 | Get parameter nodes mapped by their fully qualified names. This function traverses the graph to find all parameter | STRING |
| HIGH | torch/_functorch/_aot_autograd/fx_utils.py | 259 | Get buffer nodes mapped by their fully qualified names. This function traverses the graph to find all buffer input | STRING |
| HIGH | torch/_functorch/_aot_autograd/fx_utils.py | 287 | Get all parameter nodes from a graph as a list. You can rely on this providing the correct order of parameters you | STRING |
| HIGH | torch/_functorch/_aot_autograd/fx_utils.py | 308 | Get all buffer nodes from a graph as a list. You can rely on this providing the correct order of buffers you need | STRING |
| HIGH | torch/linalg/__init__.py | 97 | .. seealso:: :func:`torch.linalg.cholesky_ex` for a version of this operation that skips the (slow) e | STRING |
| HIGH | torch/linalg/__init__.py | 243 | .. note:: Consider using :func:`torch.linalg.solve` if possible for multiplying a matrix on the left by the in | STRING |
| HIGH | torch/linalg/__init__.py | 308 | Args: A (Tensor): tensor of shape `(*, n, n)` where `*` is zero or more batch dimensions. Keyword args: left | STRING |
| HIGH | torch/linalg/__init__.py | 426 | linalg.slogdet(A, *, out=None) -> (Tensor, Tensor) Computes the sign and natural logarithm of the absolute value of th | STRING |
| HIGH | torch/linalg/__init__.py | 504 | .. warning:: This function assumes that :attr:`A` is `diagonalizable`_ (for example, when all the eigenva | STRING |
| HIGH | torch/linalg/__init__.py | 615 | .. seealso:: :func:`torch.linalg.eig` computes the full eigenvalue decomposition. Args: A (Tensor): tens | STRING |
| HIGH | torch/linalg/__init__.py | 675 | .. note:: The eigenvalues of real symmetric or complex Hermitian matrices are always real. .. warning:: The eigenvect | STRING |
| HIGH | torch/linalg/__init__.py | 798 | .. seealso:: :func:`torch.linalg.eigh` computes the full eigenvalue decomposition. Args: A (Tensor): ten | STRING |
| HIGH | torch/linalg/__init__.py | 838 | householder_product(A, tau, *, out=None) -> Tensor Computes the first `n` columns of a product of Householder matrices | STRING |
| HIGH | torch/linalg/__init__.py | 939 | Args: A (Tensor): tensor of shape `(*, n, n)` where `*` is zero or more batch dimensions consistin | STRING |
| HIGH | torch/linalg/__init__.py | 993 | Args: A (Tensor): tensor of shape `(*, n, n)` where `*` is zero or more batch dimensions consistin | STRING |
| HIGH | torch/linalg/__init__.py | 1080 | torch.linalg.lstsq(A, B, rcond=None, *, driver=None) -> (Tensor, Tensor, Tensor, Tensor) Computes a solution to the le | STRING |
| HIGH | torch/linalg/__init__.py | 1204 | matrix_power(A, n, *, out=None) -> Tensor Computes the `n`-th power of a square matrix for an integer `n`. Supports i | STRING |
| HIGH | torch/linalg/__init__.py | 1355 | linalg.norm(input, ord=None, dim=None, keepdim=False, *, out=None, dtype=None) -> Tensor Computes a vector or matrix n | STRING |
| HIGH | torch/linalg/__init__.py | 1493 | linalg.vector_norm(x, ord=2, dim=None, keepdim=False, *, dtype=None, out=None) -> Tensor Computes a vector norm. If : | STRING |
| HIGH | torch/linalg/__init__.py | 1572 | linalg.matrix_norm(A, ord='fro', dim=(-2, -1), keepdim=False, *, dtype=None, out=None) -> Tensor Computes a matrix nor | STRING |
| HIGH | torch/linalg/__init__.py | 1736 | linalg.svd(A, full_matrices=True, *, driver=None, out=None) -> (Tensor, Tensor, Tensor) Computes the singular value de | STRING |
| HIGH | torch/linalg/__init__.py | 1907 | .. seealso:: :func:`torch.linalg.svd` computes the full singular value decomposition. Args: A (Tensor): | STRING |
| HIGH | torch/linalg/__init__.py | 1940 | linalg.cond(A, p=None, *, out=None) -> Tensor Computes the condition number of a matrix with respect to a matrix norm. | STRING |
| HIGH | torch/linalg/__init__.py | 2310 | .. seealso:: :func:`torch.linalg.solve_triangular` computes the solution of a triangular system of linear | STRING |
| HIGH | torch/linalg/__init__.py | 2474 | .. warning:: The LU decomposition is almost never unique, as often there are different permutation matrice | STRING |
| HIGH | torch/linalg/__init__.py | 2643 | lu(A, *, pivot=True, out=None) -> (Tensor, Tensor, Tensor) Computes the LU decomposition with partial pivoting of a ma | STRING |
| HIGH | torch/linalg/__init__.py | 2733 | linalg.tensorinv(A, ind=2, *, out=None) -> Tensor Computes the multiplicative inverse of :func:`torch.tensordot`. If | STRING |
| HIGH | torch/linalg/__init__.py | 2801 | linalg.tensorsolve(A, B, dims=None, *, out=None) -> Tensor Computes the solution `X` to the system `torch.tensordot(A, | STRING |
| HIGH | torch/linalg/__init__.py | 2867 | qr(A, mode='reduced', *, out=None) -> (Tensor, Tensor) Computes the QR decomposition of a matrix. Letting :math:`\mat | STRING |
| HIGH | torch/linalg/__init__.py | 3015 | Args: A (Tensor): tensor of shape `(*, m, n)` with `m >= n`, where `*` is zero or more batch dimens | STRING |
| HIGH | torch/nn/functional.py | 2226 | Sample from the Gumbel-Softmax distribution (`Link 1`_ `Link 2`_) and optionally discretize. Args: logi | STRING |
| HIGH | torch/nn/functional.py | 6497 | Args: query (Tensor): Query tensor; shape :math:`(N, ..., Hq, L, E)`. key (Tensor): Key tensor; sha | STRING |
| HIGH | torch/nn/attention/bias.py | 192 | Handles the logic for computing attention with the specified causal bias. Args: query (Ten | STRING |
| HIGH | torch/nn/attention/flex_attention.py | 1077 | Creates a BlockMask instance from key-value block information. Args: kv_num_blocks (Tensor | STRING |
| HIGH | torch/nn/attention/flex_attention.py | 1248 | Returns a new BlockMask by selecting batch, head, and Q-block rows. BlockMask indexing accepts up to t | STRING |
| HIGH | torch/nn/parallel/distributed.py | 71 | Factory method to create a BucketCapacityConfig from user inputs. Args: bucket_cap_mb: Single bucke | STRING |
| HIGH | torch/nn/parallel/distributed.py | 133 | Compute bucket size limits for initial bucketing and rebuilding. Args: static_graph: Whether the co | STRING |
| HIGH | torch/nn/utils/stateless.py | 18 | Unties all tied tensors in the module to parameters_and_buffers. This function returns a new untied_parameters | STRING |
| HIGH | torch/nn/utils/parametrize.py | 457 | Register a parametrization to a tensor in a module. Assume that ``tensor_name="weight"`` for simplicity. When acces | STRING |
| HIGH | torch/nn/utils/parametrize.py | 701 | Remove the parametrizations on a tensor in a module. - If ``leave_parametrized=True``, ``module[tensor_name]`` will | STRING |
| HIGH | torch/nn/utils/weight_norm.py | 85 | Apply weight normalization to a parameter in the given module. .. math:: \mathbf{w} = g \dfrac{\mathbf{v}} | STRING |
| HIGH | torch/nn/utils/prune.py | 590 | Compute and returns a mask for the input tensor ``t``. Starting from a base ``default_mask`` (which should be a | STRING |
| HIGH | torch/nn/utils/prune.py | 699 | Compute and returns a mask for the input tensor ``t``. Starting from a base ``default_mask`` (which should be a | STRING |
| HIGH | torch/nn/utils/prune.py | 837 | Apply pruning reparameterization without pruning any units. Applies pruning reparameterization to the tensor corres | STRING |
| 147 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tools/linter/adapters/clangtidy_linter.py | 0 | (?mx) ^ (?p<file>.*?): (?p<line>\d+): (?:(?p<column>-?\d+):)? \s(?p<severity>\s+?):? \s(?p<message>.*) \s(?p<code>\[.*\] | STRING |
| HIGH | tools/linter/adapters/mypy_linter.py | 0 | (?mx) ^ (?p<file>.*?): (?p<line>\d+): (?:(?p<column>-?\d+):)? \s(?p<severity>\s+?):? \s(?p<message>.*) \s(?p<code>\[.*\] | STRING |
| HIGH | tools/linter/adapters/pyrefly_linter.py | 0 | (?mx) ^ (?p<file>.*?): (?p<line>\d+): (?:(?p<column>-?\d+):)? \s(?p<severity>\s+?):? \s(?p<message>.*) \s(?p<code>\[.*\] | STRING |
| HIGH | .ci/pytorch/test_example_code/cnn_smoke_win_arm64.py | 0 | it's used to check basic rnn features with cuda. for example, it would throw exception if missing some components are mi | STRING |
| HIGH | .ci/pytorch/test_example_code/rnn_smoke_win_arm64.py | 0 | it's used to check basic rnn features with cuda. for example, it would throw exception if missing some components are mi | STRING |
| HIGH | .ci/pytorch/test_example_code/cnn_smoke.py | 0 | it's used to check basic rnn features with cuda. for example, it would throw exception if missing some components are mi | STRING |
| HIGH | .ci/pytorch/test_example_code/rnn_smoke.py | 0 | it's used to check basic rnn features with cuda. for example, it would throw exception if missing some components are mi | STRING |
| HIGH | test/test_transformers.py | 0 | clones the query, key, and value tensors and moves them to the specified dtype. | STRING |
| HIGH | test/inductor/test_flex_attention.py | 0 | clones the query, key, and value tensors and moves them to the specified dtype. | STRING |
| HIGH | test/inductor/test_flex_decoding.py | 0 | clones the query, key, and value tensors and moves them to the specified dtype. | STRING |
| HIGH | benchmarks/transformer/score_mod.py | 0 | clones the query, key, and value tensors and moves them to the specified dtype. | STRING |
| HIGH | test/test_dynamic_spec_make_fx.py | 0 | reset the global intvar uid counter so uids start at 0 per test. | STRING |
| HIGH | test/dynamo/test_dynamic_spec.py | 0 | reset the global intvar uid counter so uids start at 0 per test. | STRING |
| HIGH | test/export/test_dynamic_spec_export.py | 0 | reset the global intvar uid counter so uids start at 0 per test. | STRING |
| HIGH | test/inductor/test_codecache.py | 0 | verify that we can populate and hot load functions from the cache. | STRING |
| HIGH | test/inductor/test_config.py | 0 | verify that we can populate and hot load functions from the cache. | STRING |
| HIGH | test/dynamo/test_aot_autograd_cache.py | 0 | verify that we can populate and hot load functions from the cache. | STRING |
| HIGH | …ension/torch_openreg/torch_openreg/openreg/__init__.py | 0 | context-manager that changes the selected device. args: device (torch.device or int or str): device index to select. it' | STRING |
| HIGH | torch/cuda/__init__.py | 0 | context-manager that changes the selected device. args: device (torch.device or int or str): device index to select. it' | STRING |
| HIGH | torch/mtia/__init__.py | 0 | context-manager that changes the selected device. args: device (torch.device or int or str): device index to select. it' | STRING |
| HIGH | torch/xpu/__init__.py | 0 | context-manager that changes the selected device. args: device (torch.device or int or str): device index to select. it' | STRING |
| HIGH | test/nn/attention/test_fa3.py | 0 | test that torch.compile preserves tensor metadata (shape, stride, dtype). | STRING |
| HIGH | test/nn/attention/_fa_test_common.py | 0 | test that torch.compile preserves tensor metadata (shape, stride, dtype). | STRING |
| HIGH | test/nn/attention/test_fa4.py | 0 | test that torch.compile preserves tensor metadata (shape, stride, dtype). | STRING |
| HIGH | test/distributed/test_c10d_ucc.py | 0 | this unit test verifies whether the future object is passed properly using nccl backend. the hook callback function crea | STRING |
| HIGH | test/distributed/test_c10d_gloo.py | 0 | this unit test verifies whether the future object is passed properly using nccl backend. the hook callback function crea | STRING |
| HIGH | test/distributed/test_c10d_nccl.py | 0 | this unit test verifies whether the future object is passed properly using nccl backend. the hook callback function crea | STRING |
| HIGH | test/dynamo/test_aot_compile.py | 0 | \ <internal traceback> torch._dynamo.exc.unsupported: call to `torch._dynamo.graph_break()` explanation: user-inserted g | STRING |
| HIGH | test/dynamo/test_error_messages.py | 0 | \ <internal traceback> torch._dynamo.exc.unsupported: call to `torch._dynamo.graph_break()` explanation: user-inserted g | STRING |
| HIGH | test/dynamo/test_exc.py | 0 | \ <internal traceback> torch._dynamo.exc.unsupported: call to `torch._dynamo.graph_break()` explanation: user-inserted g | STRING |
| HIGH | test/quantization/eager/test_quantize_eager_qat.py | 0 | create a qat module from a float module args: `mod` a float module, either produced by torch.ao.quantization utilities o | STRING |
| HIGH | torch/ao/nn/qat/modules/linear.py | 0 | create a qat module from a float module args: `mod` a float module, either produced by torch.ao.quantization utilities o | STRING |
| HIGH | torch/ao/nn/intrinsic/qat/modules/conv_fused.py | 0 | create a qat module from a float module args: `mod` a float module, either produced by torch.ao.quantization utilities o | STRING |
| HIGH | torch/ao/nn/qat/modules/conv.py | 0 | create a qat module from a float module args: `mod` a float module, either produced by torch.ao.quantization utilities o | STRING |
| HIGH | torch/ao/nn/qat/modules/embedding_ops.py | 0 | create a qat module from a float module args: `mod` a float module, either produced by torch.ao.quantization utilities o | STRING |
| HIGH | test/cpython/v3_13/test_cmath.py | 0 | assert that floats x and y are identical, in the sense that: (1) both x and y are nans, or (2) both x and y are infiniti | STRING |
| HIGH | test/cpython/v3_13/test_complex.py | 0 | assert that floats x and y are identical, in the sense that: (1) both x and y are nans, or (2) both x and y are infiniti | STRING |
| HIGH | test/cpython/v3_13/test_float.py | 0 | assert that floats x and y are identical, in the sense that: (1) both x and y are nans, or (2) both x and y are infiniti | STRING |
| HIGH | test/jit/mydecorator.py | 0 | helper function used in test_decorator.py. we define it in a separate file on purpose to test that the names in differen | STRING |
| HIGH | test/jit/myfunction_b.py | 0 | helper function used in test_decorator.py. we define it in a separate file on purpose to test that the names in differen | STRING |
| HIGH | test/jit/myfunction_a.py | 0 | helper function used in test_decorator.py. we define it in a separate file on purpose to test that the names in differen | STRING |
| HIGH | test/functorch/test_rearrange.py | 0 | adapted from https://github.com/arogozhnikov/einops/blob/36c7bb16e57d6e57f8f3050f9e07abdf3f00469f/einops/parsing.py. mit | STRING |
| HIGH | test/functorch/test_parsing.py | 0 | adapted from https://github.com/arogozhnikov/einops/blob/36c7bb16e57d6e57f8f3050f9e07abdf3f00469f/einops/parsing.py. mit | STRING |
| HIGH | torch/_appdirs.py | 0 | adapted from https://github.com/arogozhnikov/einops/blob/36c7bb16e57d6e57f8f3050f9e07abdf3f00469f/einops/parsing.py. mit | STRING |
| HIGH | functorch/einops/_parsing.py | 0 | adapted from https://github.com/arogozhnikov/einops/blob/36c7bb16e57d6e57f8f3050f9e07abdf3f00469f/einops/parsing.py. mit | STRING |
| HIGH | test/functorch/test_codegen_mutation_epilogue.py | 0 | capture codegen artifacts from the structured trace log. | STRING |
| HIGH | test/functorch/test_codegen_backward_epilogue.py | 0 | capture codegen artifacts from the structured trace log. | STRING |
| HIGH | test/functorch/test_codegen_debug_assert.py | 0 | capture codegen artifacts from the structured trace log. | STRING |
| HIGH | test/functorch/test_codegen_output_alias.py | 0 | capture codegen artifacts from the structured trace log. | STRING |
| HIGH | test/functorch/test_codegen_dedup.py | 0 | capture codegen artifacts from the structured trace log. | STRING |
| HIGH | torch/storage.py | 0 | return a copy of this object in custom device memory. if this object is already in device memory and on the correct devi | STRING |
| HIGH | torch/_utils.py | 0 | return a copy of this object in custom device memory. if this object is already in device memory and on the correct devi | STRING |
| HIGH | torch/utils/backend_registration.py | 0 | return a copy of this object in custom device memory. if this object is already in device memory and on the correct devi | STRING |
| HIGH | torch/_export/passes/replace_with_hop_pass_util.py | 0 | split gm into sub-graph-modules using `sequential_split_and_maybe_inline_subgraphs`, and then recursively call itself on | STRING |
| HIGH | torch/_export/passes/replace_autocast_with_hop_pass.py | 0 | split gm into sub-graph-modules using `sequential_split_and_maybe_inline_subgraphs`, and then recursively call itself on | STRING |
| HIGH | torch/_export/passes/replace_set_grad_with_hop_pass.py | 0 | split gm into sub-graph-modules using `sequential_split_and_maybe_inline_subgraphs`, and then recursively call itself on | STRING |
| HIGH | torch/nn/qat/dynamic/modules/linear.py | 0 | intrinsic qat modules. this file is in the process of migration to `torch/ao/nn/intrinsic/qat`, and is kept here for com | STRING |
| HIGH | torch/nn/qat/modules/linear.py | 0 | intrinsic qat modules. this file is in the process of migration to `torch/ao/nn/intrinsic/qat`, and is kept here for com | STRING |
| HIGH | torch/nn/qat/modules/conv.py | 0 | intrinsic qat modules. this file is in the process of migration to `torch/ao/nn/intrinsic/qat`, and is kept here for com | STRING |
| HIGH | torch/nn/qat/modules/embedding_ops.py | 0 | intrinsic qat modules. this file is in the process of migration to `torch/ao/nn/intrinsic/qat`, and is kept here for com | STRING |
| 114 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | tools/nightly_hotpatch.py | 168 | # Check if the patch was applied successfully | COMMENT |
| LOW | tools/packaging/build_wheel.py | 89 | # Check if it's PyPy (skip it) | COMMENT |
| LOW | tools/packaging/build_wheel.py | 107 | # Check if this version is supported | COMMENT |
| LOW | tools/experimental/torchfuzz/ops_fuzzer.py | 71 | # Check if the operator supports any of the allowed operations | COMMENT |
| LOW | tools/experimental/torchfuzz/multi_process_fuzzer.py | 167 | # Check if output should be ignored and which pattern matched | COMMENT |
| LOW | tools/experimental/torchfuzz/operators/layout.py | 333 | # Check if there's at least one singleton dimension | COMMENT |
| LOW | tools/linter/adapters/grep_linter.py | 201 | # Check if file is too large to compute replacement | COMMENT |
| LOW | tools/linter/adapters/stable_shim_version_linter.py | 109 | # Check if this is a manual AOTI shim - only enforce versioning on new lines, | COMMENT |
| LOW | tools/linter/adapters/stable_shim_version_linter.py | 159 | # Check if this is a newly added line | COMMENT |
| LOW | tools/linter/adapters/_stable_shim_utils.py | 348 | # Check if we're entering a block comment | COMMENT |
| LOW | tools/linter/adapters/_stable_shim_utils.py | 413 | # Check if this #elif has a version condition | COMMENT |
| LOW | tools/linter/adapters/s3_init.py | 42 | # Open the file in binary mode and hash it. | COMMENT |
| LOW | tools/linter/adapters/update_s3.py | 25 | # Open the file in binary mode and hash it. | COMMENT |
| LOW | tools/linter/adapters/newlines_linter.py | 44 | # Check if file is too large | COMMENT |
| LOW | tools/linter/adapters/codespell_linter.py | 106 | # Check if file is too large | COMMENT |
| LOW | tools/amd_build/build_amd.py | 150 | # Check if the compiler is hip-clang. | COMMENT |
| LOW | tools/testing/target_determination/heuristics/utils.py | 174 | # Check if the file extension is in the docs-only set | COMMENT |
| LOW | …ols/lite_interpreter/gen_selected_mobile_ops_header.py | 87 | # Write the file selected_mobile_ops.h with optionally: | STRING |
| LOW | …ols/lite_interpreter/gen_selected_mobile_ops_header.py | 126 | # Write the file selected_mobile_ops.h with optionally: | STRING |
| LOW | tools/vendoring/quack/vendor.sh | 128 | # Set UPSTREAM_DIR to a quack checkout at $sha. With a local checkout, validate | COMMENT |
| LOW | tools/stats/import_test_stats.py | 57 | # Check if the file is new enough (see: FILE_CACHE_LIFESPAN_SECONDS). A real check | COMMENT |
| LOW | tools/stats/check_disabled_tests.py | 77 | # Check if the test is a failure | COMMENT |
| LOW | .ci/docker/common/install_torch_tpu.sh | 56 | # Check if xtrace (set -x) is enabled | COMMENT |
| LOW | .ci/pytorch/common_utils.sh | 63 | # Check if the first argument contains multiple paths separated by spaces | COMMENT |
| LOW | .ci/pytorch/common_utils.sh | 213 | # Check if the wheel has been already been built | COMMENT |
| LOW | .ci/pytorch/build.sh | 149 | # Set ROCM_ARCH to gfx906 for CI builds, if user doesn't override. | COMMENT |
| LOW | .ci/pytorch/test.sh | 186 | # Set OMP_NUM_THREADS to nproc/4 on k8s ARC runners if not already set. | COMMENT |
| LOW | .ci/pytorch/test.sh | 1453 | # Check if regression was consistent (majority of runs) | COMMENT |
| LOW | .ci/pytorch/smoke_test/smoke_test.py | 580 | # Check if NVSHMEM is compiled in current build | COMMENT |
| LOW | .ci/pytorch/smoke_test/smoke_test.py | 596 | # Check if NVSHMEM is available on current system. | COMMENT |
| LOW | .ci/pytorch/smoke_test/check_binary_symbols.py | 429 | # Check if the stable C shim exists | STRING |
| LOW | test/test_linalg.py | 9709 | # Set filenames to be unique on a per test basis | STRING |
| LOW | test/test_reductions.py | 3408 | # Check if reduction happens along the specified dim with and without keepdim. Check with | COMMENT |
| LOW | test/test_reductions.py | 3429 | # Check if function raises error on specified zero'd dimension as reduction dim. | COMMENT |
| LOW | test/test_reductions.py | 3467 | # Check if function raises error on specified zero'd dimension as reduction dim. | COMMENT |
| LOW | test/test_reductions.py | 3494 | # Check if reduction happens along the specified dimension. | COMMENT |
| LOW | test/test_bundled_images.py | 90 | # Check if fb::image_decode_to_NCHW works as expected | COMMENT |
| LOW | test/test_jit.py | 12888 | # Check if the peak sizes at most differ by an empirically obtained factor | STRING |
| LOW | test/test_view_ops.py | 1119 | # Check if result is a view of x by comparing data pointers | COMMENT |
| LOW | test/test_view_ops.py | 1121 | # Check if ref is a view of x by comparing data pointers | COMMENT |
| LOW | test/test_custom_ops.py | 333 | # Check if ret is a tuple and has exactly one and the same element | COMMENT |
| LOW | test/test_meta.py | 421 | # Check if it's a view, by testing if any of the returns have | COMMENT |
| LOW | test/test_fx_experimental.py | 920 | # Assign ops to different partitions so a single-output submodule exists. | STRING |
| LOW | test/test_modules.py | 100 | # Check if this module creates parameters or registers buffers. | COMMENT |
| LOW | test/test_modules.py | 111 | # Check if a parameter or buffer was created with a tensor not passed to the constructor. | COMMENT |
| LOW | test/test_modules.py | 269 | # Check if the inplace variant of the module gives the same result as the out of place | COMMENT |
| LOW⚡ | test/test_transformers.py | 2894 | # Check if masked rows are zero in output | COMMENT |
| LOW⚡ | test/test_transformers.py | 2901 | # Check if gradients for masked rows are zero | COMMENT |
| LOW | test/inductor/test_flex_attention.py | 7075 | # Check if kernel code contains warp specialization parameters | COMMENT |
| LOW | test/inductor/test_max_autotune.py | 6225 | # Set env to make it work in CI. | COMMENT |
| LOW | test/inductor/test_cutlass_fallback.py | 18 | # Check if CUTLASS is available | COMMENT |
| LOW | test/inductor/test_cooperative_reductions.py | 227 | # Iterate through elements for comparison | COMMENT |
| LOW | test/inductor/test_max_autotune_blackwell.py | 736 | # Set env to make it work in CI. | COMMENT |
| LOW | test/inductor/test_pad_mm.py | 681 | # Check if any bmm operations had dtype changes | COMMENT |
| LOW | test/inductor/test_ck_backend.py | 916 | # Set env to make it work in CI. | STRING |
| LOW | test/inductor/test_subgraph_choice.py | 174 | # Set env to make it work in CI. | COMMENT |
| LOW | test/inductor/test_foreach.py | 1073 | # Set requires_grad to be False to avoid mutating a leaf variable | COMMENT |
| LOW | test/inductor/test_op_dtype_prop.py | 185 | # Check if the op's output should be upcasted/downcasted. | COMMENT |
| LOW | test/inductor/test_aot_inductor_package.py | 422 | # Check if the .so file was build successfully | COMMENT |
| LOW | test/inductor/test_aot_inductor_package.py | 465 | # Check if the .so file was build successfully | COMMENT |
| 430 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/testing/test_run.py | 135 | CODE | |
| LOW | .ci/pytorch/smoke_test/smoke_test.py | 75 | CODE | |
| LOW | .ci/pytorch/smoke_test/max_autotune.py | 35 | CODE | |
| LOW | .ci/pytorch/test_example_code/cnn_smoke_win_arm64.py | 21 | CODE | |
| LOW | .ci/pytorch/test_example_code/cnn_smoke.py | 21 | CODE | |
| LOW | test/test_mps.py | 15691 | CODE | |
| LOW | test/test_mps.py | 11798 | CODE | |
| LOW | test/test_linalg.py | 4605 | CODE | |
| LOW | test/test_utils.py | 133 | CODE | |
| LOW | test/test_fx_passes.py | 155 | CODE | |
| LOW | test/test_fx_passes.py | 167 | CODE | |
| LOW | test/test_fx_passes.py | 179 | CODE | |
| LOW | test/test_fx_passes.py | 207 | CODE | |
| LOW | test/test_fx_passes.py | 806 | CODE | |
| LOW | test/test_fx_passes.py | 814 | CODE | |
| LOW | test/test_fx_passes.py | 933 | CODE | |
| LOW | test/test_fx_passes.py | 1116 | CODE | |
| LOW | test/test_static_runtime.py | 20 | CODE | |
| LOW | test/test_functionalization.py | 486 | CODE | |
| LOW | test/test_functionalization.py | 1176 | CODE | |
| LOW | test/test_decomp.py | 1621 | CODE | |
| LOW | test/test_mobile_optimizer.py | 321 | CODE | |
| LOW | test/test_xnnpack_integration.py | 1200 | CODE | |
| LOW | test/test_xnnpack_integration.py | 1215 | CODE | |
| LOW | test/test_torch.py | 3516 | CODE | |
| LOW | test/test_dynamic_shapes.py | 714 | CODE | |
| LOW | test/test_dynamic_shapes.py | 2496 | CODE | |
| LOW | test/test_dynamic_shapes.py | 4791 | CODE | |
| LOW | test/test_jit_fuser_te.py | 2693 | CODE | |
| LOW | test/test_nn.py | 2722 | CODE | |
| LOW | test/test_nn.py | 6788 | CODE | |
| LOW | test/test_jit.py | 697 | CODE | |
| LOW | test/test_jit.py | 1240 | CODE | |
| LOW | test/test_jit.py | 6589 | CODE | |
| LOW | test/test_jit.py | 6924 | CODE | |
| LOW | test/test_jit.py | 11764 | CODE | |
| LOW | test/test_jit.py | 11739 | CODE | |
| LOW | test/test_jit.py | 11753 | CODE | |
| LOW | test/test_jit.py | 15998 | CODE | |
| LOW | test/test_jit.py | 1857 | CODE | |
| LOW | test/test_custom_ops.py | 1226 | CODE | |
| LOW | test/test_custom_ops.py | 4395 | CODE | |
| LOW | test/test_torchfuzz_repros.py | 52 | CODE | |
| LOW | test/test_torchfuzz_repros.py | 93 | CODE | |
| LOW | test/test_torchfuzz_repros.py | 138 | CODE | |
| LOW | test/test_torchfuzz_repros.py | 181 | CODE | |
| LOW | test/test_torchfuzz_repros.py | 230 | CODE | |
| LOW | test/test_torchfuzz_repros.py | 264 | CODE | |
| LOW | test/test_opaque_obj_v2.py | 392 | CODE | |
| LOW | test/test_opaque_obj_v2.py | 3202 | CODE | |
| LOW | test/test_opaque_obj_v2.py | 3564 | CODE | |
| LOW | test/test_autograd.py | 8225 | CODE | |
| LOW | test/test_autograd.py | 8425 | CODE | |
| LOW | test/test_autograd.py | 9158 | CODE | |
| LOW | test/test_autograd.py | 14372 | CODE | |
| LOW | test/test_autograd.py | 17058 | CODE | |
| LOW | test/test_autograd.py | 17072 | CODE | |
| LOW | test/test_autograd.py | 17083 | CODE | |
| LOW | test/test_autograd.py | 16597 | CODE | |
| LOW | test/test_fake_tensor.py | 3284 | CODE | |
| 803 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/experimental/torchfuzz/README.md | 528 | ### Step 1: Create Operator Class | COMMENT |
| LOW | tools/experimental/torchfuzz/README.md | 564 | ### Step 2: Register Your Module | COMMENT |
| LOW | tools/experimental/torchfuzz/README.md | 581 | ### Step 3: Add to Template (Optional) | COMMENT |
| LOW | tools/experimental/torchfuzz/README.md | 598 | ### Step 4: Test Your Operator | COMMENT |
| LOW | test/test_varlen_attention.py | 365 | # NOTE: This class is currently CUDA-specific, although a significant portion of its | COMMENT |
| LOW⚡ | test/test_autograd.py | 8581 | # Step 1: Checkpointed forward with batch_size=4 → static graph G0 | COMMENT |
| LOW⚡ | test/test_autograd.py | 8585 | # Step 2: Non-checkpointed call with batch_size=7 triggers automatic | COMMENT |
| LOW⚡ | test/test_autograd.py | 8593 | # Step 3: Backward. Checkpoint recomputes fn(x) but now hits G1 | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 43 | // Step 1: Check inputs. This will throw an error if inputs are invalid, so no | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 46 | // Step 2: Create descriptors | COMMENT |
| LOW | test/cpp_extensions/cudnn_extension.cpp | 59 | // Step 3: Apply CuDNN function | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 71 | // Step 4: Destroy descriptors | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 73 | // Step 5: Return something (optional) | COMMENT |
| LOW⚡ | test/distributed/tensor/test_redistribute.py | 3161 | # Step 1: reduce_scatter on mesh dim 0 (size 2) | COMMENT |
| LOW⚡ | test/distributed/tensor/test_redistribute.py | 3165 | # Step 2: reduce_scatter on mesh dim 1 (size 3) | COMMENT |
| LOW | test/distributed/tensor/experimental/test_local_map.py | 842 | # Step 1: BuggyDPMatmul without spmd_types -> wrong W gradient. | COMMENT |
| LOW | test/distributed/tensor/experimental/test_local_map.py | 859 | # Step 2: BuggyDPMatmul with spmd_types=True -> caught. | COMMENT |
| LOW | test/distributed/tensor/experimental/test_local_map.py | 875 | # Step 3: CorrectDPMatmul with spmd_types=True -> correct gradients | COMMENT |
| LOW | test/distributed/tensor/parallel/test_tp_examples.py | 222 | # Step 3: Run test by comparing outputs from single-gpu and multi-gpu models. | COMMENT |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 305 | ) # Step 1: Initialize single-gpu models. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 308 | ) # Step 2: Setup tp model, place onto device mesh. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 311 | ) # Step 3: Setup optimizers for both models | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 433 | ) # Step 1: Initialize single-gpu models. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 436 | ) # Step 2: Setup tp model, place onto device mesh. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 439 | ) # Step 3: Setup optimizers for both models | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 442 | ) # Step 4: set `requires_grad` patterns | CODE |
| LOW | test/dynamo/test_activation_offloading.py | 596 | # Step 1: cold — allocates new pinned buffer | COMMENT |
| LOW | test/dynamo/test_activation_offloading.py | 609 | # Step 2: warm — reuses pooled buffer | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7737 | # Step 1: Compile a static shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7746 | # Step 2: Compile a dynamic shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7756 | # Step 3: Run with Step 1's inputs | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7779 | # Step 1: Compile a static shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7788 | # Step 2: Compile a dynamic shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7798 | # Step 3: Run with Step 1's inputs | COMMENT |
| LOW | test/profiler/test_profiler.py | 3887 | # Step 3: under the patch, RECORD -> DEVICE_STOPPED fires stop_trace + | COMMENT |
| LOW | .claude/skills/aoti-debug/triton-index-out-of-bounds.md | 31 | ## Step 1: Collect AOTI Package | COMMENT |
| LOW | .claude/skills/aoti-debug/triton-index-out-of-bounds.md | 43 | ## Step 2: Locate the Failing Kernel in C++ Wrapper | COMMENT |
| LOW | .claude/skills/aoti-debug/triton-index-out-of-bounds.md | 72 | ## Step 3: Understand the Kernel Logic | COMMENT |
| LOW | .claude/skills/aoti-debug/triton-index-out-of-bounds.md | 99 | ## Step 4: Identify the Dynamic Shape Variable | COMMENT |
| LOW | .claude/skills/aoti-debug/triton-index-out-of-bounds.md | 135 | ## Step 5: Trace Back to Model Input | COMMENT |
| LOW | .claude/skills/aoti-debug/triton-index-out-of-bounds.md | 160 | ## Step 6: Map to Model Code | COMMENT |
| LOW | .claude/skills/aoti-debug/SKILL.md | 93 | ### Step 1: Sanity Checks | COMMENT |
| LOW | .claude/skills/aoti-debug/SKILL.md | 107 | ### Step 2: Pinpoint the CUDA IMA | COMMENT |
| LOW | .claude/skills/aoti-debug/SKILL.md | 121 | ### Step 3: Identify Problematic Kernels with Intermediate Value Debugger | COMMENT |
| LOW | .claude/skills/metal-kernel/SKILL.md | 24 | ## Step 1: Update native_functions.yaml | COMMENT |
| LOW | .claude/skills/metal-kernel/SKILL.md | 89 | ## Step 2: Implement Metal Kernel | COMMENT |
| LOW | .claude/skills/metal-kernel/SKILL.md | 236 | ## Step 3: Implement Host-Side Stub | COMMENT |
| LOW | .claude/skills/metal-kernel/SKILL.md | 283 | ## Step 4: Compile | COMMENT |
| LOW | .claude/skills/pyrefly-type-coverage/SKILL.md | 14 | ### Step 1: Remove file-level type-check suppressions | COMMENT |
| LOW | .claude/skills/pyrefly-type-coverage/SKILL.md | 26 | ### Step 2: Add a sub-config entry to `pyrefly.toml` | COMMENT |
| LOW | .claude/skills/pyrefly-type-coverage/SKILL.md | 45 | ### Step 3: Run pyrefly | COMMENT |
| LOW | .claude/skills/pyrefly-type-coverage/SKILL.md | 70 | ### Step 4: Add annotations | COMMENT |
| LOW | .claude/skills/pyrefly-type-coverage/SKILL.md | 200 | ### Step 5: Iterate | COMMENT |
| LOW | .claude/skills/pyrefly-type-coverage/SKILL.md | 209 | ### Step 6: Lint | COMMENT |
| LOW | .claude/skills/pyrefly-type-coverage/SKILL.md | 220 | ### Step 7: Test | COMMENT |
| LOW⚡ | .claude/skills/pt2-bug-basher/SKILL.md | 260 | # Step 1: Generate the minifier launcher | COMMENT |
| LOW⚡ | .claude/skills/pt2-bug-basher/SKILL.md | 263 | # Step 2: Run the minifier | COMMENT |
| LOW⚡ | .claude/skills/pt2-bug-basher/SKILL.md | 266 | # Step 3: Run the minimized repro | COMMENT |
| LOW | .claude/skills/at-dispatch-v2/SKILL.md | 44 | ### Step 1: Add the Dispatch_v2.h include | COMMENT |
| LOW | .claude/skills/at-dispatch-v2/SKILL.md | 55 | ### Step 2: Identify the old dispatch pattern | COMMENT |
| 314 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | test/test_fx_passes.py | 583 | if node.op not in ("placeholder", "output") | CODE |
| LOW⚡ | test/test_fx_passes.py | 589 | node.name for node in fused_submodule.graph.nodes if node.op == "placeholder" | CODE |
| LOW | test/test_fx_passes.py | 552 | if node.op not in ("placeholder", "output") | CODE |
| LOW | test/test_fx_passes.py | 558 | node.name for node in fused_submodule.graph.nodes if node.op == "placeholder" | CODE |
| LOW | test/test_fx_passes.py | 619 | node.name for node in fused_submodule.graph.nodes if node.op == "placeholder" | CODE |
| LOW | test/test_fx_passes.py | 650 | node.name for node in fused_submodule.graph.nodes if node.op == "placeholder" | CODE |
| LOW | test/test_fx_passes.py | 1187 | if not test_case.match_placeholder and node.op == "placeholder": | CODE |
| LOW⚡ | test/test_proxy_tensor.py | 1199 | if node.op == "placeholder" | CODE |
| LOW | test/test_proxy_tensor.py | 1241 | placeholders = [node for node in gm.graph.nodes if node.op == "placeholder"] | CODE |
| LOW | test/test_proxy_tensor.py | 1914 | meta_inp = _get_node(fx_g, lambda x: x.op == 'placeholder') | CODE |
| LOW | test/test_dynamic_shapes.py | 968 | if n.op == "placeholder" and "val" in n.meta | CODE |
| LOW | test/test_dynamic_shapes.py | 7347 | if node.op == "placeholder": | CODE |
| LOW | test/test_opaque_obj_v2.py | 4283 | if n.op == "placeholder" and is_opaque_node(n) | CODE |
| LOW | test/test_opaque_obj_v2.py | 4306 | placeholders = [n for n in captured_gm.graph.nodes if n.op == "placeholder"] | CODE |
| LOW | test/test_opaque_obj_v2.py | 4320 | restored_phs = [n for n in restored_gm.graph.nodes if n.op == "placeholder"] | CODE |
| LOW | test/test_opaque_obj_v2.py | 4364 | placeholders = [n for n in gm.graph.nodes if n.op == "placeholder"] | CODE |
| LOW⚡ | test/test_fx_experimental.py | 1384 | self.assertEqual(nodes[0].op, "placeholder") | CODE |
| LOW | test/test_fx_experimental.py | 345 | if node.op not in {"output", "placeholder", "get_attr"}: | CODE |
| LOW | test/test_fx_experimental.py | 403 | if node.op not in {"output", "placeholder", "get_attr"}: | CODE |
| LOW | test/test_fx_experimental.py | 468 | if node.op not in {"placeholder", "get_attr", "output"}: | CODE |
| LOW | test/test_fx_experimental.py | 1166 | if node.op in ("placeholder", "get_attr", "output"): | CODE |
| LOW | test/test_fx_experimental.py | 1221 | if node.op in ("placeholder", "get_attr", "output"): | CODE |
| LOW | test/test_fx_experimental.py | 1337 | if node.op in ("placeholder", "get_attr", "output"): | CODE |
| LOW | test/test_fx_experimental.py | 1362 | if node.op in ("placeholder", "get_attr", "output"): | CODE |
| LOW⚡ | test/test_fx_experimental.py | 1019 | self.assertEqual(nodes[0].op, "placeholder") | STRING |
| LOW⚡ | test/test_fx_experimental.py | 1594 | if node.op == "placeholder": | STRING |
| LOW | test/test_fx_experimental.py | 1153 | elif node.op == "placeholder": | STRING |
| LOW | test/test_fx_experimental.py | 1609 | ("placeholder", "x"), | STRING |
| LOW | test/test_fx_experimental.py | 1638 | ("placeholder", "x"), | STRING |
| LOW⚡ | test/test_dynamic_spec_make_fx.py | 35 | if node.op != "placeholder": | CODE |
| LOW⚡ | test/test_fx.py | 2242 | if node.op == "placeholder": | CODE |
| LOW⚡ | test/test_fx.py | 2257 | node.meta.get("val") for node in gm.graph.nodes if node.op == "placeholder" | CODE |
| LOW⚡ | test/test_fx.py | 2677 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW⚡ | test/test_fx.py | 2687 | x : torch.fx.Node = graph.create_node('placeholder', 'x') | CODE |
| LOW⚡ | test/test_fx.py | 2696 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 915 | if node.op in {"placeholder", "output"}: | CODE |
| LOW | test/test_fx.py | 964 | a: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 1046 | if n.op == "placeholder": | CODE |
| LOW | test/test_fx.py | 1111 | placeholder_nodes.append(graph.create_node("placeholder", name)) | CODE |
| LOW | test/test_fx.py | 1463 | FileCheck().check("x").check("placeholder").check("y").check( | CODE |
| LOW | test/test_fx.py | 1464 | "placeholder" | CODE |
| LOW | test/test_fx.py | 1476 | a: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 1966 | a: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2075 | a: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2126 | "placeholder", | CODE |
| LOW | test/test_fx.py | 2169 | if node.op == "placeholder": | CODE |
| LOW | test/test_fx.py | 2734 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2799 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2800 | y: torch.fx.Node = graph.create_node("placeholder", "y") | CODE |
| LOW | test/test_fx.py | 2816 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2834 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2835 | y: torch.fx.Node = graph.create_node("placeholder", "y") | CODE |
| LOW | test/test_fx.py | 2851 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2852 | y: torch.fx.Node = graph.create_node("placeholder", "y") | CODE |
| LOW | test/test_fx.py | 2902 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2920 | x: torch.fx.Node = graph.create_node("placeholder", "x") | CODE |
| LOW | test/test_fx.py | 2991 | if not (input_node and input_node.op == "placeholder"): | CODE |
| LOW | test/test_fx.py | 2993 | f"Expected input_node to exist and have op='placeholder', " | CODE |
| LOW | test/test_fx.py | 3348 | if node.op == "placeholder": | CODE |
| LOW | test/test_fx.py | 3484 | if node.op == "placeholder" or node.op == "output": | CODE |
| 671 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/optional_submodules.py | 71 | # Otherwise just call top-level function of choice | COMMENT |
| MEDIUM | tools/experimental/torchfuzz/visualize_graph.py | 79 | # Create comprehensive label | COMMENT |
| LOW | tools/experimental/torchfuzz/operators/layout.py | 69 | # For single-dim input, just use the exact element count | COMMENT |
| LOW | tools/experimental/torchfuzz/operators/layout.py | 150 | # For single-dim input, just use the exact element count | COMMENT |
| MEDIUM⚡ | tools/test/test_native_aot_tools.py | 99 | # name, which this by-path test harness can't provide.) | COMMENT |
| MEDIUM | tools/test/test_native_aot_tools.py | 184 | # real compile, which this harness cannot do). | COMMENT |
| LOW | tools/linter/adapters/gb_registry_linter.py | 210 | # Empty registry, just add all entries | COMMENT |
| MEDIUM | tools/autograd/gen_python_functions.py | 596 | # find and leverage the original ATen signature (to which it delegates | COMMENT |
| MEDIUM | tools/autograd/gen_autograd_functions.py | 196 | # `grads` in order to utilize and apply the existing derivative definitions | COMMENT |
| MEDIUM | tools/autograd/gen_autograd_functions.py | 684 | # note(crcrpar): [nuanced return type of out-of-place foreach functions] | STRING |
| MEDIUM | tools/autograd/gen_variable_type.py | 427 | # pertaining to its output: | COMMENT |
| MEDIUM | tools/autograd/gen_variable_type.py | 1567 | # See Note [nuanced return type of out-of-place foreach functions] | COMMENT |
| LOW | tools/autograd/gen_variable_type.py | 2124 | # View ops create fw_grad that already is a view of the base's fw_grad so just use that | COMMENT |
| LOW | tools/autograd/gen_trace_type.py | 180 | # For inplace: just add the input to the end to confirm with the JIT schema | COMMENT |
| MEDIUM | tools/alerts/create_alerts.py | 182 | # Creates a Dict of Job Name -> [JobData]. Essentially a Column in HUD | STRING |
| LOW | tools/code_coverage/package/oss/init.py | 128 | # if this argument is specified, just return itself | COMMENT |
| MEDIUM | tools/stats/upload_test_stats.py | 67 | # To do this, we leverage an implementation detail of how we write out | COMMENT |
| LOW | .ci/docker/manywheel/s390_scripts/build.sh | 51 | # So let's just use the same one pip and everyone uses | COMMENT |
| LOW | .ci/docker/common/install_nccl.sh | 6 | # so we can just use the default NCCL version. | COMMENT |
| LOW | .ci/manywheel/build_common.sh | 176 | # Note - just use whichever python we happen to be on | COMMENT |
| LOW | .ci/pytorch/common_utils.sh | 158 | # There won't be nvidia-smi in nogpu tests, so just set TORCH_CUDA_ARCH_LIST to the default | COMMENT |
| MEDIUM | .ci/pytorch/run_tests.sh | 7 | # Essentially runs pytorch/test/run_test.py, but keeps track of which tests to | COMMENT |
| MEDIUM | .ci/pytorch/test_fa3_abi_stable.sh | 22 | # To be as robust against upstream changes as possible, minimally only stage | COMMENT |
| MEDIUM | test/test_mps.py | 7130 | # Essentially remained over integral types should rely on integers ops | COMMENT |
| LOW | test/test_mps.py | 15927 | # fail, so just set CWD to this script's directory | STRING |
| LOW | test/test_mps.py | 15972 | # fail, so just set CWD to this script's directory | STRING |
| LOW | test/test_python_dispatch.py | 2292 | e.contiguous() # this will just return the original TensorImpl since is_contiguous = True | CODE |
| LOW | test/test_import_stats.py | 7 | # time is greater than a certain threshold, but for now we just use them | COMMENT |
| LOW | test/test_tensor_creation_ops.py | 2649 | # instead of repeating this calculation, we just use empty_strided which does the same | COMMENT |
| MEDIUM⚡ | test/test_nn.py | 14648 | # Probability-target counterpart of the scalar harness legs: | COMMENT |
| LOW | test/test_pytree.py | 913 | # fail, so just set CWD to this script's directory | STRING |
| MEDIUM | test/test_custom_ops.py | 1499 | # Not comprehensive (it doesn't need to be), just a check that our mechanism works | COMMENT |
| LOW⚡ | test/test_stateless.py | 911 | # fail, so just set CWD to this script's directory | STRING |
| LOW⚡ | test/test_stateless.py | 935 | # fail, so just set CWD to this script's directory | STRING |
| MEDIUM | test/test_jit_llga_fuser.py | 61 | # so we disable AMP for JIT & leverage eager-mode AMP. | COMMENT |
| LOW | test/test_cuda.py | 7905 | # fail, so just set CWD to this script's directory | COMMENT |
| LOW | test/test_cuda.py | 3904 | # fail, so just set CWD to this script's directory | STRING |
| MEDIUM | test/run_test.py | 448 | # in that they utilize the usual python test infrastructure. | STRING |
| LOW | test/run_test.py | 549 | # to just return successfully here | STRING |
| MEDIUM | test/test_dlpack.py | 843 | # Run the comprehensive C++ test | STRING |
| MEDIUM | test/test_meta.py | 791 | # useful for more comprehensive testing e.g., as seen in | STRING |
| MEDIUM | test/test_binary_ufuncs.py | 2070 | # in a binary operation in conjunction with a Tensor on all | COMMENT |
| MEDIUM | test/test_autograd.py | 3029 | # Definitely pick one tensor to add | COMMENT |
| LOW | test/test_autograd.py | 5134 | # This op will just return grad_output two times in backward | COMMENT |
| LOW | test/test_autograd.py | 12612 | # fail, so just set CWD to this script's directory | STRING |
| MEDIUM | test/test_nestedtensor.py | 9031 | # OpInfo-based NJT tests. These tests utilize an NJT-specific op_db generated from the standard | COMMENT |
| MEDIUM | test/test_scaled_matmul_cuda.py | 1472 | # by cosine similarity (robust to kernel differences) rather than element-wise. | COMMENT |
| LOW | test/test_testing.py | 2793 | # fail, so just set CWD to this script's directory | COMMENT |
| MEDIUM | test/inductor/test_deterministic.py | 136 | # Halving sweep, matching what the benchmark harness does. | COMMENT |
| LOW⚡ | test/inductor/test_compiled_autograd.py | 186 | # fail, so just set CWD to this script's directory | COMMENT |
| LOW | test/inductor/test_triton_kernels.py | 5871 | # we can just pass the function 'f' for dynamo | STRING |
| LOW | test/inductor/test_triton_kernels.py | 6155 | # we can just pass the function 'f' for dynamo | STRING |
| MEDIUM | test/inductor/test_max_autotune_blackwell.py | 305 | # Fundamentally we should fix the triton_blackwell_ws_persistent_device_tma template | COMMENT |
| MEDIUM | test/inductor/test_cpu_repro.py | 534 | # The cpp_wrapper C-shim can't utilize the Python error API, so error | COMMENT |
| MEDIUM⚡ | test/inductor/test_compile_to_python.py | 426 | # profiling harness. ``def call(`` matches both the flat ``def call(args)`` and the | COMMENT |
| MEDIUM | test/inductor/test_compile_to_python.py | 178 | # than an Inductor profiling harness: none of these debug entry points appear. | COMMENT |
| MEDIUM | test/inductor/test_control_flow.py | 61 | # Simple robust loss: abs values + small constant to avoid inf/nan | COMMENT |
| MEDIUM | test/inductor/test_torchinductor.py | 17503 | # odd number on purpose to trigger comprehensive padding | COMMENT |
| MEDIUM⚡ | test/inductor/test_torchinductor_opinfo_properties.py | 761 | # Use reference_inputs for more comprehensive test coverage | COMMENT |
| MEDIUM⚡ | test/inductor/test_torchinductor_opinfo_properties.py | 760 | """Get sample inputs from OpInfo using reference_inputs for comprehensive coverage.""" | STRING |
| 254 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | test/test_autograd.py | 8581 | # Step 1: Checkpointed forward with batch_size=4 → static graph G0 | COMMENT |
| LOW⚡ | test/test_autograd.py | 8585 | # Step 2: Non-checkpointed call with batch_size=7 triggers automatic | COMMENT |
| LOW⚡ | test/test_autograd.py | 8593 | # Step 3: Backward. Checkpoint recomputes fn(x) but now hits G1 | COMMENT |
| LOW | test/test_testing.py | 1048 | # We need to check if this test is still needed or if this behavior is now enabled by default. | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 43 | // Step 1: Check inputs. This will throw an error if inputs are invalid, so no | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 46 | // Step 2: Create descriptors | COMMENT |
| LOW | test/cpp_extensions/cudnn_extension.cpp | 59 | // Step 3: Apply CuDNN function | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 71 | // Step 4: Destroy descriptors | COMMENT |
| LOW⚡ | test/cpp_extensions/cudnn_extension.cpp | 73 | // Step 5: Return something (optional) | COMMENT |
| LOW⚡ | test/distributed/tensor/test_redistribute.py | 3161 | # Step 1: reduce_scatter on mesh dim 0 (size 2) | COMMENT |
| LOW⚡ | test/distributed/tensor/test_redistribute.py | 3165 | # Step 2: reduce_scatter on mesh dim 1 (size 3) | COMMENT |
| LOW | test/distributed/tensor/experimental/test_local_map.py | 842 | # Step 1: BuggyDPMatmul without spmd_types -> wrong W gradient. | COMMENT |
| LOW | test/distributed/tensor/experimental/test_local_map.py | 859 | # Step 2: BuggyDPMatmul with spmd_types=True -> caught. | COMMENT |
| LOW | test/distributed/tensor/experimental/test_local_map.py | 875 | # Step 3: CorrectDPMatmul with spmd_types=True -> correct gradients | COMMENT |
| LOW | test/distributed/tensor/parallel/test_tp_examples.py | 222 | # Step 3: Run test by comparing outputs from single-gpu and multi-gpu models. | COMMENT |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 305 | ) # Step 1: Initialize single-gpu models. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 308 | ) # Step 2: Setup tp model, place onto device mesh. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 311 | ) # Step 3: Setup optimizers for both models | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 433 | ) # Step 1: Initialize single-gpu models. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 436 | ) # Step 2: Setup tp model, place onto device mesh. | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 439 | ) # Step 3: Setup optimizers for both models | CODE |
| LOW⚡ | test/distributed/tensor/parallel/test_tp_examples.py | 442 | ) # Step 4: set `requires_grad` patterns | CODE |
| LOW | test/dynamo/test_activation_offloading.py | 596 | # Step 1: cold — allocates new pinned buffer | COMMENT |
| LOW | test/dynamo/test_activation_offloading.py | 609 | # Step 2: warm — reuses pooled buffer | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7737 | # Step 1: Compile a static shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7746 | # Step 2: Compile a dynamic shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7756 | # Step 3: Run with Step 1's inputs | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7779 | # Step 1: Compile a static shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7788 | # Step 2: Compile a dynamic shapes graph | COMMENT |
| LOW⚡ | test/dynamo/test_repros.py | 7798 | # Step 3: Run with Step 1's inputs | COMMENT |
| LOW | test/profiler/test_profiler.py | 3887 | # Step 3: under the patch, RECORD -> DEVICE_STOPPED fires stop_trace + | COMMENT |
| LOW | torch/_custom_ops.py | 61 | >>> # Step 1: define the custom op. | CODE |
| LOW | torch/_custom_ops.py | 72 | >>> # Step 2: Register an implementation for various PyTorch subsystems | CODE |
| LOW | torch/_custom_ops.py | 144 | >>> # Step 1: define the custom op. | CODE |
| LOW | torch/_custom_ops.py | 155 | >>> # Step 2: Register an implementation for various PyTorch subsystems | CODE |
| LOW | torch/_ops.py | 421 | # Step 1: dispatch on any user TorchDispatchModes | COMMENT |
| LOW | torch/_ops.py | 449 | # Step 2: dispatch on any subclasses | COMMENT |
| LOW | torch/_ops.py | 682 | # When we are unsetting a mode, we need to check if there is | COMMENT |
| LOW | torch/_ops.py | 732 | # When we are setting a mode, we need to check if there is | COMMENT |
| LOW | torch/csrc/dynamo/init.cpp | 362 | // even if __next__ is not defined, so we need to check if the method is truly | COMMENT |
| LOW | torch/csrc/utils/python_arg_parser.cpp | 566 | // Step 1: Try to dispatch based on the mode stack, *ignoring* infra | COMMENT |
| LOW | torch/csrc/utils/python_arg_parser.cpp | 616 | // Step 1: Try to dispatch on any user TorchDispatchModes (including infra | COMMENT |
| LOW | torch/csrc/utils/python_arg_parser.cpp | 627 | // Step 2: Try to dispatch based on any user subclasses, | COMMENT |
| LOW | torch/csrc/jit/tensorexpr/loopnest.cpp | 1975 | // We need to check if fusing the loops results in a loop-carried dependence. | COMMENT |
| LOW | torch/csrc/lazy/ts_backend/ts_eager_fallback.cpp | 222 | // Step 1: Convert all non-eager tensor inputs into eager tensors and put them | COMMENT |
| LOW⚡ | torch/csrc/lazy/ts_backend/ts_eager_fallback.cpp | 254 | // Step 2: Call the underlying eager implementation of the operator | COMMENT |
| LOW⚡ | torch/csrc/lazy/ts_backend/ts_eager_fallback.cpp | 257 | // Step 3: We need to take special care to handle mutable aliases properly: | COMMENT |
| LOW | torch/csrc/lazy/ts_backend/ts_eager_fallback.cpp | 268 | // Step 4: Convert any eager output tensors back to the original input device. | COMMENT |
| LOW | torch/_functorch/eager_transforms.py | 661 | # Step 1: Construct grad_outputs by splitting the standard basis | COMMENT |
| LOW | torch/_functorch/eager_transforms.py | 778 | # Step 2: The returned jacobian is one big tensor per input. In this step, | COMMENT |
| LOW | torch/_functorch/eager_transforms.py | 792 | # Step 3: Right now, `jacobian` is a List[List[Tensor]]. | COMMENT |
| LOW | torch/_functorch/_aot_autograd/functional_utils.py | 413 | # For outputs aliasing inputs, we need to check if the requires-gradness has changed. | COMMENT |
| LOW | torch/_functorch/_aot_autograd/functional_utils.py | 415 | # For outputs aliasing inputs, we need to check if the dtype has changed. | COMMENT |
| LOW⚡ | …rch/_functorch/_aot_autograd/graph_capture_wrappers.py | 1341 | # Step 1: wrap tensor inputs into subclasses if necessary | COMMENT |
| LOW⚡ | …rch/_functorch/_aot_autograd/graph_capture_wrappers.py | 1346 | # Step 2: call the inner function, with our (maybe subclass) inputs | COMMENT |
| LOW | …rch/_functorch/_aot_autograd/graph_capture_wrappers.py | 1379 | # Step 3: Unwrap any subclass outputs back into dense tensors | COMMENT |
| LOW⚡ | …ctorch/_activation_offloading/activation_offloading.py | 782 | # Step 1: Identify reload patterns | COMMENT |
| LOW⚡ | …ctorch/_activation_offloading/activation_offloading.py | 787 | # Step 2: Reorder nodes by directly manipulating the graph | COMMENT |
| LOW⚡ | …ctorch/_activation_offloading/activation_offloading.py | 806 | # Step 1: Decide which nodes to offload and mark them | COMMENT |
| LOW⚡ | …ctorch/_activation_offloading/activation_offloading.py | 815 | # Step 2: Add offload and reload nodes to the graphs | COMMENT |
| 212 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | .ci/macwheel/build.sh | 5 | # repair_wheel.py) are non-orchestrating stages that stay version-agnostic. | COMMENT |
| MEDIUM | .ci/macwheel/build_all_macos_wheels.sh | 8 | # Per-Python orchestration (env setup, deps, build, delocate) lives in | COMMENT |
| MEDIUM | …ofiler/profiler_global_callback_stale_session_test.cpp | 12 | #include <torch/csrc/profiler/orchestration/observer.h> | COMMENT |
| MEDIUM | test/profiler/test_cupti_monitor.py | 549 | # The window orchestration: keys each launch by correlation id -> exec graph id (from a | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 307 | # AROUND the orchestration (graph_compile._aot_stage2c_make_inference_function), | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 309 | # orchestration -- dedup(orchestration(inner)) -- so the orchestration sees the | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 338 | # orchestration. It collapses the aliased inputs into a synthetic base BEFORE the | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 339 | # orchestration runs, so it must be composed OUTSIDE it (same requirement as dedup): | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 341 | # which inverted the chain to orchestration(synthetic_base(inner))). | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 418 | # so the inner compile MUST run under autocast for the orchestration to emit | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 420 | # baked in; the orchestration then disables autocast at runtime to keep the dense | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 898 | # exactly one forward orchestration wrapper and rejects an empty capture rather than | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 908 | # The inlined orchestration hoists its globals to module scope; a hoisted name that | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 1021 | # A valid forward orchestration wrapper whose codegen'd signature matches what the | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 1023 | # broken sibling so the SIBLING is what trips the guard, not a missing orchestration. | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 1035 | # front (before chain wiring) even when paired with a valid forward orchestration: | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 1095 | # target origin (the last orchestration's), so a foreign-origin wrapper must be | COMMENT |
| MEDIUM⚡ | test/functorch/test_compile_to_python.py | 1096 | # dropped from the emitted source while the target orchestration still composes. | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 145 | # _inner_call, and the AOTAutograd orchestration is inlined as a real def that the | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 283 | # (the _alias_fn / gen_alias_from_base path, which the orchestration closes over | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 397 | # those wrap the orchestration rather than the inner call -- see the dedup / | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 471 | # The orchestration is spliced as a real top-level ``def _runtime_wrapper`` that the | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 483 | # the orchestration is invoked with contextlib.nullcontext + a no-op in those two | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 525 | # that token is the orchestration's first parameter and appears in every composed | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 856 | # The generated call invokes the orchestration positionally, so a changed signature | COMMENT |
| MEDIUM | test/functorch/test_compile_to_python.py | 1080 | # Two orchestration wrappers sharing one origin_id (so the origin filter keeps both) | COMMENT |
| MEDIUM | torch/csrc/autograd/profiler_python.cpp | 29 | #include <torch/csrc/profiler/orchestration/python_tracer.h> | COMMENT |
| MEDIUM | torch/csrc/autograd/profiler_kineto.cpp | 20 | #include <torch/csrc/profiler/orchestration/observer.h> | COMMENT |
| MEDIUM | torch/csrc/mtia/profiler/MTIAMemoryProfiler.h | 2 | #include <torch/csrc/profiler/orchestration/python_tracer.h> | COMMENT |
| MEDIUM | torch/csrc/profiler/collection.cpp | 2 | #include <torch/csrc/profiler/orchestration/vulkan.h> | COMMENT |
| MEDIUM | torch/csrc/profiler/api.h | 3 | #include <torch/csrc/profiler/orchestration/observer.h> | COMMENT |
| MEDIUM | torch/csrc/profiler/collection.h | 21 | #include <torch/csrc/profiler/orchestration/python_tracer.h> | COMMENT |
| MEDIUM | torch/csrc/profiler/orchestration/python_tracer.cpp | 1 | #include <torch/csrc/profiler/orchestration/python_tracer.h> | COMMENT |
| MEDIUM | torch/csrc/profiler/orchestration/observer.cpp | 1 | #include <torch/csrc/profiler/orchestration/observer.h> | COMMENT |
| MEDIUM | torch/csrc/profiler/orchestration/vulkan.cpp | 1 | #include <torch/csrc/profiler/orchestration/vulkan.h> | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/runtime_wrappers.py | 988 | # The orchestration closes over these two codegen'd epilogue functions directly | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 321 | # caught by the orchestration count/wiring guards below (a second orchestration trips | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 323 | # target is the origin of the LAST captured orchestration wrapper: a foreign lowering | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 324 | # appends its orchestration before the outer one finishes, so the final orchestration | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 422 | # This is what lets the composer tell the inner call apart from the orchestration's | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 425 | # wrapping the inner call, composed INSIDE the orchestration) are distinguished from | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 427 | # orchestration in graph_compile._aot_stage2c_make_inference_function). | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 430 | # Name of the single-arg adapter emitted over the orchestration; the innermost outer | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 431 | # wrapper closes over the orchestration's outer closure and is wired to this name. | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 434 | # The orchestration's outer closure is the object outer wrappers wrap. The composer | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 435 | # never captures it directly (the captured orchestration ``fn`` is the inner | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 621 | # The single-arg adapter over the orchestration (emitted only when outer wrappers | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 623 | # orchestration's positional (_compiled_fn_, _first_ctx_, _on_before_call_, args) | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 624 | # signature to that. When there are no outer wrappers the orchestration is invoked | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 705 | # ``call`` invokes the orchestration directly when nothing wraps it; when outer | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 707 | # adapter, then the orchestration) -- so the orchestration-substitution comment lives | COMMENT |
| MEDIUM⚡ | torch/_functorch/_aot_autograd/to_standalone_python.py | 708 | # on whichever site actually invokes the orchestration. | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 93 | # calls the link it wraps -- the orchestration's outer closure for the innermost outer | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 94 | # wrapper), plus the orchestration's output-alias and mutation epilogue helpers, which | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 108 | # _replay_aliases_ = _alias_fn # (3) orchestration's ref | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 223 | # codegen'd runtime wrappers, not reimplemented: each (the orchestration and any chain | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 246 | # An OUTER wrapper (dedup / synthetic base) closes over the orchestration's outer | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 248 | # path, so wire it to the single-arg orchestration entry adapter the composer emits. | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 360 | # The generated ``call`` invokes the orchestration POSITIONALLY by its own name (see | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/to_standalone_python.py | 408 | # ``_compiled_fn_``). The orchestration takes its inner as a call-time arg, not a | COMMENT |
| 18 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/test/heuristics/test_heuristics.py | 270 | def helper(test_file: str, changed_file: str) -> bool: | CODE |
| LOW⚡ | test/test_mps.py | 4061 | def helper(binary_op): | CODE |
| LOW⚡ | test/test_mps.py | 5038 | def helper(dtype): | CODE |
| LOW⚡ | test/test_mps.py | 5049 | def helper(dtype): | CODE |
| LOW | test/test_mps.py | 257 | def helper(model_cpu, model_mps, dtype, iterations, batch_size, atol=3e-4, rtol=1e-5): | CODE |
| LOW | test/test_mps.py | 321 | def helper(model, optimizer, scaler, device, input, target, apply_grad_transform=False): | CODE |
| LOW | test/test_mps.py | 1105 | def helper(val, shape, dtype): | CODE |
| LOW | test/test_mps.py | 1321 | def helper(dtype): | CODE |
| LOW | test/test_mps.py | 1740 | def helper(input_shape, batch1_shape, batch2_shape): | CODE |
| LOW | test/test_mps.py | 1816 | def helper(bias_shape): | CODE |
| LOW | test/test_mps.py | 2153 | def helper(shape, ks, padding=0, dilation=1, ceil_mode=False, return_indices=False, test_ties=False): | CODE |
| LOW | test/test_mps.py | 2227 | def helper(size, memory_format): | CODE |
| LOW | test/test_mps.py | 2253 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 2362 | def helper(shape, channels_last=False): | CODE |
| LOW | test/test_mps.py | 2379 | def helper(shape, eps=1, momentum=0.1, wts=False, training=False, channels_last=False, | CODE |
| LOW | test/test_mps.py | 2955 | def helper(input_shape, normalized_shape, eps=1e-05, elementwise_affine=True, dtype=torch.float32, non_contiguou | CODE |
| LOW | test/test_mps.py | 3085 | def helper(shape, eps=1, momentum=0.1, wts=False, channels_last=False, track_running_stats=True, test_module=Fal | CODE |
| LOW | test/test_mps.py | 3262 | def helper(dim, layer='linear', dtype=torch.float32): | CODE |
| LOW | test/test_mps.py | 3332 | def helper(input_shape, wt_shape, | CODE |
| LOW | test/test_mps.py | 3433 | def helper(input_shape, wt_shape, | CODE |
| LOW | test/test_mps.py | 3501 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 3526 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 3550 | def helper(threshold, value, num_elems, inplace=False, requires_grad=True): | CODE |
| LOW | test/test_mps.py | 3577 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 3610 | def helper(shape, value, xtype=torch.float32, ytype=None, ztype=None): | CODE |
| LOW | test/test_mps.py | 3647 | def helper(shape, value): | CODE |
| LOW | test/test_mps.py | 3673 | def helper(shape, value): | CODE |
| LOW | test/test_mps.py | 3863 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 3983 | def helper(x_tensor, y_func, z_func, r_func=None): | CODE |
| LOW | test/test_mps.py | 4095 | def helper(operator): | CODE |
| LOW | test/test_mps.py | 4187 | def helper(shape, repeats): | CODE |
| LOW | test/test_mps.py | 4286 | def helper(shape, dtype=torch.float32, num_repeats=torch.Tensor(), dim=None): | CODE |
| LOW | test/test_mps.py | 4311 | def helper(dtype): | CODE |
| LOW | test/test_mps.py | 4411 | def helper(data, to_dtype): | CODE |
| LOW | test/test_mps.py | 4429 | def helper(data, to_dtype): | CODE |
| LOW | test/test_mps.py | 4542 | def helper(dtype): | CODE |
| LOW | test/test_mps.py | 4599 | def helper(x, return_inverse, return_counts): | CODE |
| LOW | test/test_mps.py | 4628 | def helper(x, dim, return_inverse, return_counts): | CODE |
| LOW | test/test_mps.py | 4820 | def helper(src_shape, dst_shape, src_dtype, dst_dtype): | CODE |
| LOW | test/test_mps.py | 4958 | def helper(dtype, noncontiguous, dim): | CODE |
| LOW | test/test_mps.py | 5003 | def helper(dtype): | CODE |
| LOW | test/test_mps.py | 5021 | def helper(dtype): | CODE |
| LOW | test/test_mps.py | 5131 | def helper(shape, dtype): | CODE |
| LOW | test/test_mps.py | 5331 | def helper(n, c): | CODE |
| LOW | test/test_mps.py | 5348 | def helper(shape, reduction): | CODE |
| LOW | test/test_mps.py | 5377 | def helper(shape, reduction): | CODE |
| LOW | test/test_mps.py | 5447 | def helper(shape, reduction): | CODE |
| LOW | test/test_mps.py | 5721 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5734 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5745 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5758 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5769 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5782 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5793 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5806 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5817 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5830 | def helper(shape): | CODE |
| LOW | test/test_mps.py | 5859 | def helper(n, c, h, w, reduction_type, dtype=torch.float32): | CODE |
| LOW | test/test_mps.py | 6053 | def helper(n, c, h, w, dtype=torch.float32): | CODE |
| LOW | test/test_mps.py | 6218 | def helper(shape): | CODE |
| 201 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | test/test_datapipe.py | 3741 | CODE | |
| MEDIUM | test/test_functionalization.py | 770 | CODE | |
| MEDIUM | test/test_functionalization.py | 771 | CODE | |
| MEDIUM | test/test_functionalization.py | 772 | CODE | |
| MEDIUM | test/test_functionalization.py | 773 | CODE | |
| MEDIUM | test/test_jit.py | 14084 | CODE | |
| MEDIUM | test/test_jit.py | 9934 | CODE | |
| MEDIUM | test/test_jit_fuser.py | 734 | CODE | |
| MEDIUM | test/test_jit_fuser.py | 735 | CODE | |
| MEDIUM | test/test_jit_fuser.py | 737 | CODE | |
| MEDIUM | test/test_jit_fuser.py | 739 | CODE | |
| MEDIUM | test/test_jit_fuser.py | 742 | CODE | |
| MEDIUM | test/test_jit_fuser.py | 746 | CODE | |
| MEDIUM | test/test_autograd.py | 16480 | CODE | |
| MEDIUM | test/test_fx.py | 5428 | CODE | |
| MEDIUM | test/test_fx.py | 4002 | CODE | |
| MEDIUM | test/inductor/test_ordered_set.py | 25 | CODE | |
| MEDIUM | test/inductor/test_ordered_set.py | 976 | CODE | |
| MEDIUM | test/onnx/test_utility_funs.py | 109 | CODE | |
| MEDIUM | test/dynamo/test_misc.py | 10499 | CODE | |
| MEDIUM | test/dynamo/test_misc.py | 13833 | CODE | |
| MEDIUM | test/dynamo/test_generator.py | 1030 | CODE | |
| MEDIUM | test/dynamo/test_generator.py | 1041 | CODE | |
| MEDIUM | test/dynamo/test_generator.py | 1515 | CODE | |
| MEDIUM | test/dynamo/test_exceptions.py | 737 | CODE | |
| MEDIUM | test/dynamo/test_guard_manager.py | 988 | CODE | |
| MEDIUM | test/quantization/fx/test_quantize_fx.py | 2564 | CODE | |
| MEDIUM | test/quantization/fx/test_quantize_fx.py | 2566 | CODE | |
| MEDIUM | test/quantization/fx/test_quantize_fx.py | 2569 | CODE | |
| MEDIUM | test/cpython/v3_13/test_itertools.py | 2723 | CODE | |
| MEDIUM | test/cpython/v3_13/test_generators.py | 566 | CODE | |
| MEDIUM | test/cpython/v3_13/test_generators.py | 577 | CODE | |
| MEDIUM | test/cpython/v3_13/test_exceptions.py | 1836 | CODE | |
| MEDIUM | test/cpython/v3_13/test_set.py | 77 | CODE | |
| MEDIUM | test/cpython/v3_13/test_set.py | 1109 | CODE | |
| MEDIUM | test/cpython/v3_13/test_deque.py | 40 | CODE | |
| MEDIUM | test/cpython/v3_13/test_builtin.py | 1288 | CODE | |
| MEDIUM | test/jit/test_backends.py | 595 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 501 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 987 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 988 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 989 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 992 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1003 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1004 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1005 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1007 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1013 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1014 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1151 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1152 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1207 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1209 | CODE | |
| MEDIUM | test/torch_np/numpy_tests/core/test_indexing.py | 1210 | CODE | |
| MEDIUM | test/functorch/test_vmap.py | 1004 | CODE | |
| MEDIUM | test/functorch/test_vmap.py | 1007 | CODE | |
| MEDIUM | test/functorch/test_vmap.py | 1010 | CODE | |
| MEDIUM | test/functorch/test_vmap.py | 1013 | CODE | |
| MEDIUM | test/functorch/test_vmap.py | 1017 | CODE | |
| MEDIUM | test/functorch/test_aotdispatch.py | 367 | CODE | |
| 19 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tools/autograd/derivatives.yaml | 37 | # If None of the output is differentiable, you can also add the function | COMMENT |
| LOW | test/test_jit.py | 455 | # you are deleting `constants.pkl`, make sure to update `torch.serialization.load` | COMMENT |
| LOW | test/test_jit.py | 2536 | # NB: We make sure to pass in a batch with a different max sequence | STRING |
| MEDIUM | test/test_fx.py | 4255 | # Used to test that you can use your own placeholder class | COMMENT |
| LOW | test/inductor/test_triton_kernels.py | 6395 | # so we wanna make sure to correctly handle edge cases with tricky line breaks | STRING |
| MEDIUM | test/inductor/test_profiler.py | 83 | # Debugging tips: you can add prof.export_chrome_trace("test.json") inline in | COMMENT |
| LOW | test/dynamo/test_repros.py | 1637 | # NB: When you remove the expectedFailure, don't forget to | COMMENT |
| LOW | test/dynamo/test_repros.py | 1637 | # NB: When you remove the expectedFailure, don't forget to | COMMENT |
| LOW | test/quantization/fx/test_model_report_fx.py | 1958 | # if they passed in fusion parameter, make sure to test that | COMMENT |
| MEDIUM | test/ao/sparsity/test_structured_sparsifier.py | 1006 | # Manually set the filter weights for demonstration purposes | COMMENT |
| LOW | .claude/skills/triaging-issues/labels.json | 416 | "description": "Optimization passes written in FX (don't forget to select a more specific label)" | CODE |
| MEDIUM | torch/torch_version.py | 38 | # Ideally this should work for most cases by attempting to group | COMMENT |
| LOW | torch/headeronly/core/ScalarType.h | 341 | // Don't forget to extend this when adding new QInt types | COMMENT |
| MEDIUM | torch/csrc/jit/backends/backend_debug_info.h | 51 | * don't serialize this info as part of LoweredModule as mentioned earlier. | COMMENT |
| MEDIUM | torch/_functorch/_aot_autograd/graph_compile.py | 999 | # Step 1) - Get the `joint_hop_gm`. As mentioned earlier, the | COMMENT |
| LOW⚡ | torch/_subclasses/meta_utils.py | 962 | # Need to make sure to resize the meta storage too. | COMMENT |
| LOW | torch/nn/utils/parametrizations.py | 135 | # We always make sure to copy Q in every path | COMMENT |
| MEDIUM | torch/nn/modules/loss.py | 2215 | >>> S_min = 10 # Minimum target length, for demonstration purposes | CODE |
| LOW | torch/nn/modules/activation.py | 1363 | # When lifting this restriction, don't forget to either | STRING |
| LOW | torch/nn/modules/activation.py | 1363 | # When lifting this restriction, don't forget to either | STRING |
| LOW | …nnx/_internal/torchscript_exporter/symbolic_opset11.py | 980 | # make sure to cast other to self's type | COMMENT |
| LOW | …nnx/_internal/torchscript_exporter/symbolic_opset11.py | 1004 | # make sure to cast other to self's type | COMMENT |
| LOW | …onnx/_internal/torchscript_exporter/symbolic_opset9.py | 2184 | # make sure to cast other to self's type | COMMENT |
| LOW | …onnx/_internal/torchscript_exporter/symbolic_opset9.py | 2222 | # make sure to cast other to self's type | COMMENT |
| LOW | …nnx/_internal/torchscript_exporter/symbolic_opset12.py | 77 | # make sure to cast other to self's type | COMMENT |
| LOW | torch/distributed/elastic/rendezvous/etcd_rendezvous.py | 886 | # make sure you don't call any long-blocking methods that do not | COMMENT |
| LOW | torch/distributed/tensor/_dtensor_spec.py | 448 | # use, where we make sure to update the hash when the `tensor_meta` | COMMENT |
| LOW | torch/fx/graph.py | 2494 | # When generating Python code, we need to make sure to name things | COMMENT |
| MEDIUM | torch/fx/experimental/symbolic_shapes.py | 1542 | # (1) If you can use a higher level combinator prefer using those instead, they are definitely safe (modulo short-circui | COMMENT |
| LOW⚡ | torch/fx/experimental/proxy_tensor.py | 753 | # from_real_tensor to create fake values (don't forget to | COMMENT |
| LOW⚡ | torch/fx/experimental/proxy_tensor.py | 753 | # from_real_tensor to create fake values (don't forget to | COMMENT |
| LOW | torch/fx/passes/reinplace.py | 764 | # When that happens, we need to make sure to keep our | STRING |
| MEDIUM | torch/optim/lr_scheduler.py | 368 | >>> # Alternatively, you can use a single lambda function for all groups. | CODE |
| LOW | torch/_inductor/cpp_builder.py | 2703 | # When we build remotely, we need to make sure to carefully copy any files | COMMENT |
| LOW | torch/_inductor/compile_fx_ext.py | 644 | # Since our metrics were gathered in a subprocess make sure to add them | COMMENT |
| LOW | torch/_inductor/codegen/common.py | 1355 | # NB: if you add a new special function, don't forget to update | COMMENT |
| LOW | torch/_inductor/codegen/common.py | 1355 | # NB: if you add a new special function, don't forget to update | COMMENT |
| LOW | torch/utils/_python_dispatch.py | 956 | # Next: we need to make sure to return inputs directly, if the output is a mutable alias (e.g. add_()). | COMMENT |
| LOW | torch/utils/hipify/hipify_python.py | 734 | # make sure to check the end-of-word marker present | COMMENT |
| LOW | torch/utils/data/sampler.py | 73 | # implementing a `__len__` method. In such cases, we must make sure to not | STRING |
| MEDIUM | torch/_dynamo/backends/onnxrt.py | 34 | # " (4) If it returns `True`, then you can use `onnxrt` backend.\n" | COMMENT |
| MEDIUM | benchmarks/dynamo/common.py | 1036 | # Hypothetically you can use this from other places, but it's currently | COMMENT |
| LOW | .github/scripts/test_trymerge.py | 8 | # GraphQL queries in trymerge.py, please make sure to delete `gql_mocks.json` | COMMENT |
| LOW | aten/src/ATen/OpaqueTensorImpl.h | 49 | // unnecessary; don't forget to change that if needed! | COMMENT |
| LOW | aten/src/ATen/SparseTensorImpl.cpp | 50 | // unnecessary; don't forget to change that if needed! | COMMENT |
| MEDIUM | torchgen/api/types/types_base.py | 179 | # which you can use to figure out if a given Binding is appropriate | COMMENT |
| LOW | c10/core/StorageImpl.h | 106 | // unnecessary; don't forget to change that if needed! | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | test/test_mps.py | 10182 | # TODO: Add tests for data types | COMMENT |
| LOW | test/test_dataloader.py | 2589 | # TODO: Fix test for 'loader_kill' that would cause running out of shared memory. | COMMENT |
| LOW | test/test_reductions.py | 1526 | # TODO: consider refactoring with bincount test | COMMENT |
| LOW | test/test_sparse.py | 5149 | # TODO: implement batch support in _convert_indices_from_csr_to_coo | COMMENT |
| LOW | test/test_ops.py | 695 | # TODO: consider adding some tolerance to this comparison | COMMENT |
| LOW | test/test_ops.py | 2252 | # TODO: add tests for `R->C` functions | COMMENT |
| LOW | …composable/test_composability/test_2d_composability.py | 509 | # TODO: Add tests when using gradient_as_bucket_view and static_graph for DDP. | COMMENT |
| LOW | test/distributed/tensor/test_matrix_ops.py | 918 | # TODO: Add test cases where is_causal=False and an attention mask is provided. | COMMENT |
| LOW | test/quantization/core/test_quantized_module.py | 1896 | # TODO: add tests for conv and linear | STRING |
| LOW | test/cpp/api/support.h | 116 | // TODO: implement abs on CharTensor (int8) | COMMENT |
| LOW | test/jit/test_tracer.py | 644 | # TODO: implement | COMMENT |
| LOW | test/torch_np/test_reductions.py | 364 | # TODO: add tests with np.int32(3) etc, when implemented | COMMENT |
| LOW | test/torch_np/test_reductions.py | 524 | # TODO: add tests with np.int32(3) etc, when implemented | COMMENT |
| LOW | test/functorch/test_ops.py | 1020 | # TODO: implement batching rule | COMMENT |
| LOW | test/functorch/test_ops.py | 1134 | # TODO: implement batching rule | COMMENT |
| LOW | test/functorch/test_ops.py | 1289 | # TODO: implement batching rule | COMMENT |
| LOW | test/functorch/test_ops.py | 1492 | # TODO: implement batching rule | COMMENT |
| LOW | test/functorch/test_ops.py | 1604 | # TODO: implement batching rule | COMMENT |
| LOW | test/functorch/test_ops.py | 1986 | # TODO: implement batching rule | COMMENT |
| LOW | test/functorch/test_vmap.py | 4570 | # TODO: implement batching rule | COMMENT |
| LOW | test/functorch/test_vmap.py | 4638 | # TODO: implement batching rule | COMMENT |
| LOW | torch/_prims/__init__.py | 400 | # TODO: implement dtype validation here, too, or on the corresponding refs | COMMENT |
| LOW | torch/csrc/Storage.cpp | 365 | // TODO: implement equality | COMMENT |
| LOW | torch/csrc/inductor/aoti_torch/shim_common.cpp | 466 | // TODO: implement a more efficient version instead of calling into aten | COMMENT |
| LOW | torch/csrc/inductor/aoti_torch/shim_common.cpp | 865 | // TODO: implement a more efficient version instead of calling into aten | COMMENT |
| LOW | torch/csrc/inductor/aoti_torch/shim_common.cpp | 936 | // TODO: implement a more efficient version instead of calling into aten | COMMENT |
| LOW | torch/csrc/inductor/aoti_torch/shim_common.cpp | 954 | // TODO: implement a more efficient version instead of calling into aten | COMMENT |
| LOW | torch/csrc/inductor/aoti_torch/shim_common.cpp | 977 | // TODO: implement a more efficient version instead of calling into aten | COMMENT |
| LOW | torch/csrc/jit/mobile/nnc/aot_compiler.cpp | 152 | // TODO: implement prealloc optimization and fill in temp_sizes | COMMENT |
| LOW | torch/distributed/elastic/rendezvous/etcd_rendezvous.py | 949 | # TODO: implement timeout | COMMENT |
| LOW⚡ | torch/distributed/_shard/sharded_optim/api.py | 82 | # TODO: implement state_dict | STRING |
| LOW⚡ | torch/distributed/_shard/sharded_optim/api.py | 92 | # TODO: implement load_state_dict | STRING |
| LOW⚡ | torch/distributed/_shard/sharded_optim/api.py | 99 | # TODO: implement add_param_group | COMMENT |
| LOW | torch/fx/traceback.py | 77 | # TODO: add logging to tlparse | COMMENT |
| LOW | torch/_prims_common/wrappers.py | 55 | # TODO: implement ref.cast with an option to enforce safe casting | COMMENT |
| LOW | torch/masked/_ops.py | 846 | # TODO: implement sparse CSR specific where operator for efficiency | COMMENT |
| LOW | torch/_inductor/utils.py | 4402 | # TODO: implement V3_BACKENDS_TUPLE | COMMENT |
| LOW | torch/utils/model_dump/__init__.py | 356 | # TODO: handle errors here and just ignore the file? | COMMENT |
| LOW | torch/ao/quantization/fx/_lower_to_native_backend.py | 399 | # TODO: add tests for lowering these ops | COMMENT |
| LOW | torch/ao/quantization/fx/fuse.py | 119 | # TODO: add validation that root_node is a module and has the same type | COMMENT |
| LOW | torch/_refs/__init__.py | 1865 | # TODO: consider refactoring this with add impl | COMMENT |
| LOW | torch/_refs/special/__init__.py | 235 | # TODO: add docstring | COMMENT |
| LOW | aten/src/ATen/LegacyBatchedTensorImpl.cpp | 85 | // TODO: implement proper contiguity on batched tensor, then put | COMMENT |
| LOW | aten/src/ATen/TensorIndexing.h | 211 | // TODO: implement negative step | COMMENT |
| LOW | aten/src/ATen/native/Embedding.cpp | 81 | // TODO: implement scale_grad_by_freq | COMMENT |
| LOW | aten/src/ATen/native/LinearAlgebra.cpp | 564 | // TODO: implement _out variant avoiding copy and using already allocated storage directly | COMMENT |
| LOW | aten/src/ATen/native/LinearAlgebra.cpp | 3399 | // TODO: implement _out variant avoiding copy and using already allocated storage directly | COMMENT |
| LOW | aten/src/ATen/native/LinearAlgebra.cpp | 3448 | // TODO: implement _out variant avoiding copy and using already allocated storage directly | COMMENT |
| LOW | aten/src/ATen/native/quantized/cudnn/Conv.cpp | 299 | // TODO: implement this for groups > 1; should be straightforward since we're only padding a single dimension | COMMENT |
| LOW | aten/src/ATen/native/quantized/cudnn/ConvPrepack.cpp | 80 | // TODO: implement this for groups > 1 | COMMENT |
| LOW | aten/src/ATen/functorch/BatchedTensorImpl.cpp | 127 | // TODO: implement proper contiguity on batched tensor, then put | COMMENT |
| LOW | torchgen/model.py | 1935 | # TODO: implement a proper parser if this gets more ugly | COMMENT |
| LOW | c10/core/TensorImpl.h | 1332 | // TODO: implement layout() as native function/method so that | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | test/test_decomp.py | 1087 | # explicit clearing is necessary as I will create a fresh mode | COMMENT |
| HIGH⚡ | test/test_fx_experimental.py | 1030 | # Based on the code from - transformers/models/granitemoe/modeling_granitemoe.py | STRING |
| HIGH | test/inductor/test_max_autotune.py | 1325 | # In this example, AOTInductor's JIT-compile will fuse(buf1, buf2) due | COMMENT |
| HIGH | test/dynamo/test_repros.py | 4994 | # In this example, there were 4950 guards (roughly (# tensors) ^ 2 // 2), | COMMENT |
| HIGH | test/jit/test_freezing.py | 988 | # In this example, 'a' is not mutated. However, we do not track which sub | COMMENT |
| HIGH | torch/_dynamo/variables/higher_order_ops.py | 883 | # used. In this example, the capture graph looks like: | COMMENT |
| HIGH | torch/_dynamo/variables/higher_order_ops.py | 1897 | # In this example, list `z` is in outer scope and gets appended | COMMENT |
| HIGH | torch/_dynamo/variables/builtin.py | 806 | # In this example, A.__radd__() is not called -> nothing is printed, because | COMMENT |
| HIGH | torch/profiler/profiler.py | 989 | # In this example with wait=1, warmup=1, active=2, repeat=1, | STRING |
| HIGH | …er_guide/torch_compiler/torch.compiler_aot_inductor.md | 62 | # [Note] In this example we directly feed the exported module to aoti_compile_and_package. | COMMENT |
| HIGH | aten/src/ATen/native/mps/operations/RnnOps.mm | 845 | // In this implementation these two biases are added together and used further. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/vendoring/quack/vendor.sh | 8 | # Usage: | COMMENT |
| LOW | torch/library.h | 961 | /// Example usage: | COMMENT |
| LOW | torch/library.h | 1038 | /// Example usage: | COMMENT |
| LOW | torch/csrc/api/include/torch/python.h | 198 | /// Example usage: | COMMENT |
| LOW | torch/csrc/api/include/torch/python.h | 234 | /// Example usage: | COMMENT |
| LOW | torch/csrc/api/include/torch/nn/modules/common.h | 6 | /// Example usage: | COMMENT |
| LOW | torch/nn/attention/flex_attention.py | 49 | # Usage: | COMMENT |
| LOW | torch/_inductor/config.py | 177 | # Example usage: | COMMENT |
| LOW | torch/_inductor/fuzzer.py | 534 | # Example usage: | STRING |
| LOW | torch/testing/_internal/common_device_type.py | 1155 | # Usage: | COMMENT |
| LOW | torch/testing/_internal/common_device_type.py | 1167 | # Usage: | COMMENT |
| LOW | torch/testing/_internal/common_device_type.py | 1443 | # Example usage: | COMMENT |
| LOW | torch/_dynamo/variables/higher_order_ops.py | 151 | # Example usage: | COMMENT |
| LOW | third_party/perfetto/update.sh | 8 | # Usage: | COMMENT |
| LOW | aten/src/ATen/TensorMeta.h | 17 | // Example usage: | COMMENT |
| LOW | aten/src/ATen/TensorMeta.h | 49 | // Example usage: | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | test/test_view_ops.py | 1132 | # Test various configurations - one per line as requested | COMMENT |
| HIGH | test/quantization/fx/test_quantize_fx.py | 8938 | # reshape will be quantized to fp16 as requested by this qconfig | COMMENT |
| HIGH | c10/util/Backtrace.cpp | 226 | // Skip as many frames as requested. | COMMENT |