AIMET is a library that provides advanced quantization and compression techniques for trained neural network models.
This report summarizes scanner evidence for qualcomm/aimet, a Python project with 2,726 GitHub stars. SynthScan examined 341,215 lines of code across 1171 source files, recording 5989 pattern matches distributed across 26 syntactic categories. The legacy adjusted pattern signal is 27.2 (Elevated pattern signal); it is not an authorship estimate.
Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.
Use each vector to guide manual verification of the underlying findings.
Can maintainers explain intent, assumptions, and trade-offs? 4192 finding(s).
Are imports declared and package/API names verified? 640 finding(s).
Are wrappers, guards, and repeated structures necessary? 4453 finding(s).
Which changes need focused human review? 5989 finding(s).
This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.
Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.
This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.
The scanner identified 5989 distinct pattern matches across 26 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. 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 configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/dependencies.py | 46 | def _get_pip_reqs_for_framework( | CODE |
| LOW | GenAILab/bench/metrics.py | 134 | def _compute_loss_from_logits( | CODE |
| LOW | GenAILab/bench/yaml_config_parser.py | 611 | def get_required_export_adaptations(cls, model_type: str) -> list[str]: | CODE |
| LOW | GenAILab/bench/recipe_cache.py | 620 | def _serialize_dataset_config(dataset_config: dict) -> dict: | CODE |
| LOW | GenAILab/bench/datasets.py | 280 | def _format_question_and_answer( | CODE |
| LOW | GenAILab/bench/datasets.py | 419 | def _format_question_and_answer( | CODE |
| LOW | GenAILab/bench/datasets.py | 495 | def assemble_fewshot_question(formatted_question: str, subject: str): | CODE |
| LOW | GenAILab/bench/datasets.py | 571 | def _format_question_and_answer( | CODE |
| LOW | GenAILab/bench/datasets.py | 1516 | def _build_transcription_inputs(self, waveform): | CODE |
| LOW | GenAILab/bench/recipe_chain.py | 163 | def apply_pre_quantization_chain( | CODE |
| LOW | GenAILab/bench/precision.py | 189 | def ensure_component_defaults(self, component: str) -> None: | CODE |
| LOW | GenAILab/bench/profiler.py | 44 | def convert_gpu_meter_to_dict( | CODE |
| LOW | GenAILab/bench/profiler.py | 84 | def convert_metric_result_to_dict( | CODE |
| LOW | GenAILab/bench/profiler.py | 191 | def _aggregate_resource_utilization( | CODE |
| LOW | GenAILab/bench/profiler.py | 218 | def _serialize_component_stats( | CODE |
| LOW | GenAILab/bench/profiler.py | 348 | def dict_to_postgres_csv_json_field(d): | CODE |
| LOW | GenAILab/bench/torch/test_genai.py | 316 | def _last_calibration_for_onnx(steps): | CODE |
| LOW | GenAILab/tests/unit/conftest.py | 29 | def _permissive_register_model(cls, model_type): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_topology_handoff.py | 43 | def test_spinquant_uses_supplied_topology(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_topology_handoff.py | 58 | def test_adascale_uses_supplied_topology(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_topology_handoff.py | 81 | def test_neither_recipe_analyzes_topology_itself(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_topology_handoff.py | 125 | def test_pre_sim_chain_forwards_topology(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_topology_handoff.py | 136 | def test_on_sim_chain_forwards_topology(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_topology_handoff.py | 162 | def test_torch_spinquant_absorbs_topology(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_topology_handoff.py | 193 | def test_topology_is_a_fixture_not_a_recipe_knob(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 19 | def test_cuda_device_with_index(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 27 | def test_cuda_device_without_index(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 35 | def test_sets_and_restores_attribute(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 46 | def test_creates_and_removes_new_attribute(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 58 | def test_restores_class_level_attribute(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 72 | def test_ties_past_key_value_quantizers(self, tmp_path): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 121 | def test_disables_activation_quantizers(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 143 | def test_float_precision_skips_tying(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 154 | def test_int_precision_ties_and_sets_symmetric(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_generator_utils.py | 50 | def test_fp_mode_disables_and_restores_quantizers(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_generator_utils.py | 78 | def test_fp_mode_covers_visual_model(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_generator_utils.py | 116 | def test_wraps_backbone_in_torch_onnx_interface(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_generator_utils.py | 143 | def test_vlm_wraps_both_models(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_generator_utils.py | 194 | def test_reassembles_list_outputs(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_generator_utils.py | 216 | def test_no_list_outputs_passthrough(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_audio_component.py | 81 | def test_audio_sim_is_wired_to_audio_model_kwarg(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 112 | def test_fp_mode_covers_audio_model(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 161 | def test_present_components_in_canonical_order(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 167 | def test_round_trip_with_audio(self, tmp_path): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 184 | def test_round_trip_without_audio_writes_no_audio_dir(self, tmp_path): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 201 | def test_legacy_entry_without_audio_dir_still_loads(self, tmp_path): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_audio_component.py | 229 | def _run_instantiate_quantsim(model_cls, entry, precision): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 260 | def instantiate_position_processor(cls): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 265 | def test_builds_audio_sim_and_no_visual_sim(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 281 | def test_no_stray_visual_precision_block(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_audio_component.py | 298 | def test_declared_component_with_no_graph_yields_no_sim(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_quantsim_instantiation.py | 61 | def test_default_precision_when_none(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_quantsim_instantiation.py | 68 | def test_calls_set_lm_head_precision(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_quantsim_instantiation.py | 74 | def test_calls_resolve_kv_cache(self): | CODE |
| LOW⚡ | GenAILab/tests/unit/onnx/test_quantsim_instantiation.py | 80 | def test_calls_block_granularity(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_instantiation.py | 95 | def test_returns_sim_collection_with_config(self): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_analysis_passes.py | 77 | def test_every_present_component_truncated_during_evaluate(build_calls): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_analysis_passes.py | 97 | def test_rebuild_during_evaluate_reinstruments(build_calls): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_analysis_passes.py | 117 | def test_restored_clean_when_evaluate_raises(build_calls): | CODE |
| LOW | GenAILab/tests/unit/onnx/test_analysis_passes.py | 131 | def test_restored_clean_when_instrumenting_a_later_sim_fails(build_calls, monkeypatch): | CODE |
| 3199 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | CMakeLists.txt | 122 | # ------------------------------- | COMMENT |
| MEDIUM | CMakeLists.txt | 124 | # ------------------------------- | COMMENT |
| MEDIUM | CMakeLists.txt | 176 | # ------------------------------- | COMMENT |
| MEDIUM | CMakeLists.txt | 178 | # ------------------------------- | COMMENT |
| MEDIUM | CMakeLists.txt | 209 | # ------------------------------- | COMMENT |
| MEDIUM | CMakeLists.txt | 211 | # ------------------------------- | COMMENT |
| MEDIUM | CMakeLists.txt | 247 | # ------------------------------- | COMMENT |
| MEDIUM | CMakeLists.txt | 249 | # ------------------------------- | COMMENT |
| MEDIUM | packaging/INSTALL.txt | 1 | # ================================================================================== | COMMENT |
| MEDIUM | packaging/INSTALL.txt | 3 | # ================================================================================== | COMMENT |
| MEDIUM | packaging/envsetup.sh | 7 | #==================================== | COMMENT |
| MEDIUM | packaging/envsetup.sh | 9 | #=================================== | COMMENT |
| MEDIUM | GenAILab/__main__.py | 173 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/__main__.py | 175 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/__main__.py | 283 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/__main__.py | 285 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/__main__.py | 409 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/__main__.py | 411 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | GenAILab/bench/metrics.py | 603 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | GenAILab/bench/metrics.py | 605 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 532 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 534 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 671 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 673 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 799 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 801 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 884 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 886 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 1407 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 1409 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 2226 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/metrics.py | 2228 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | GenAILab/bench/yaml_config_parser.py | 542 | # ======================== | COMMENT |
| MEDIUM⚡ | GenAILab/bench/yaml_config_parser.py | 544 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 202 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 204 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 241 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 243 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 285 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 287 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 317 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 319 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 410 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 412 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 663 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/yaml_config_parser.py | 665 | # ======================== | COMMENT |
| MEDIUM | GenAILab/bench/datasets.py | 1350 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/datasets.py | 1352 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | GenAILab/bench/eval_context.py | 58 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/eval_context.py | 60 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/eval_context.py | 80 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/eval_context.py | 82 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/fp_cache.py | 35 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/fp_cache.py | 37 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/fp_cache.py | 104 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/fp_cache.py | 106 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/model_cache.py | 89 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/model_cache.py | 91 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/model_cache.py | 286 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | GenAILab/bench/model_cache.py | 288 | # ------------------------------------------------------------------ | COMMENT |
| 520 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | ModelOptimizations/PyModelOptimizations/_quant_enums.py | 0 | shared quantization enums used by cython modules. this module provides intenum classes that are shared between _libpymo | STRING |
| HIGH | …gExtensions/onnx/src/python/aimet_onnx/_quant_enums.py | 0 | shared quantization enums used by cython modules. this module provides intenum classes that are shared between _libpymo | STRING |
| HIGH | …nsions/common/src/python/_aimet_common/_quant_enums.py | 0 | shared quantization enums used by cython modules. this module provides intenum classes that are shared between _libpymo | STRING |
| HIGH | Docs/snippets/onnx/apply_litemp.py | 0 | returns calibration and evaluation data-loader for imagenet dataset from provided path | STRING |
| HIGH | Docs/snippets/onnx/apply_quantsim.py | 0 | returns calibration and evaluation data-loader for imagenet dataset from provided path | STRING |
| HIGH | Docs/snippets/torch/apply_quantsim.py | 0 | returns calibration and evaluation data-loader for imagenet dataset from provided path | STRING |
| HIGH | …acy/torch_code_examples/model_preparer_code_example.py | 0 | model that uses functional modules instead of nn.modules. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …Extensions/onnx/test/python/models/models_for_tests.py | 0 | model that uses functional modules instead of nn.modules. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …ningExtensions/torch/test/python/models/test_models.py | 0 | model that uses functional modules instead of nn.modules. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …Extensions/torch/test/python/v2/models_/test_models.py | 0 | model that uses functional modules instead of nn.modules. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …s/torch/src/python/aimet_torch/examples/test_models.py | 0 | model that uses functional modules instead of nn.modules. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …ngExtensions/torch/test/python/test_model_validator.py | 0 | model that uses functional modules instead of nn.modules. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …acy/torch_code_examples/model_preparer_code_example.py | 0 | model that uses single relu instances multiple times in the forward. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …Extensions/onnx/test/python/models/models_for_tests.py | 0 | model that uses single relu instances multiple times in the forward. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …ningExtensions/torch/test/python/models/test_models.py | 0 | model that uses single relu instances multiple times in the forward. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …Extensions/torch/test/python/v2/models_/test_models.py | 0 | model that uses single relu instances multiple times in the forward. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …s/torch/src/python/aimet_torch/examples/test_models.py | 0 | model that uses single relu instances multiple times in the forward. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …cy/torch_code_examples/model_validator_code_example.py | 0 | model that reuses a relu module. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …Extensions/onnx/test/python/models/models_for_tests.py | 0 | model that reuses a relu module. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …ningExtensions/torch/test/python/models/test_models.py | 0 | model that reuses a relu module. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …Extensions/torch/test/python/v2/models_/test_models.py | 0 | model that reuses a relu module. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | …s/torch/src/python/aimet_torch/examples/test_models.py | 0 | model that reuses a relu module. expects input of shape (1, 3, 32, 32) | STRING |
| HIGH | Examples/onnx/utils/image_net_evaluator.py | 0 | :param images_dir: the path to the data directory :param image_size: the length of the image :param batch_size: the batc | STRING |
| HIGH | Examples/torch/utils/image_net_trainer.py | 0 | :param images_dir: the path to the data directory :param image_size: the length of the image :param batch_size: the batc | STRING |
| HIGH | Examples/torch/utils/image_net_evaluator.py | 0 | :param images_dir: the path to the data directory :param image_size: the length of the image :param batch_size: the batc | STRING |
| HIGH | Examples/torch/utils/image_net_evaluator.py | 0 | evaluate the specified model using the specified number of samples from the validation set. :param model: the model to b | STRING |
| HIGH | Examples/torch/quantization/range_learning.py | 0 | evaluate the specified model using the specified number of samples from the validation set. :param model: the model to b | STRING |
| HIGH | …ples/torch/quantization/quantization_aware_training.py | 0 | evaluate the specified model using the specified number of samples from the validation set. :param model: the model to b | STRING |
| HIGH | Examples/torch/quantization/adaround.py | 0 | evaluate the specified model using the specified number of samples from the validation set. :param model: the model to b | STRING |
| HIGH | Examples/torch/quantization/range_learning.py | 0 | provides apis for model quantization using evaluation and finetuning. | STRING |
| HIGH | …ples/torch/quantization/quantization_aware_training.py | 0 | provides apis for model quantization using evaluation and finetuning. | STRING |
| HIGH | Examples/torch/quantization/adaround.py | 0 | provides apis for model quantization using evaluation and finetuning. | STRING |
| HIGH | Examples/torch/quantization/range_learning.py | 0 | finetunes the model. the implemtation provided here is just an example, provide your own implementation if needed. :para | STRING |
| HIGH | Examples/torch/compression/weight_svd.py | 0 | finetunes the model. the implemtation provided here is just an example, provide your own implementation if needed. :para | STRING |
| HIGH | Examples/torch/compression/channel_pruning.py | 0 | finetunes the model. the implemtation provided here is just an example, provide your own implementation if needed. :para | STRING |
| HIGH | Examples/torch/compression/spatial_svd_cp.py | 0 | finetunes the model. the implemtation provided here is just an example, provide your own implementation if needed. :para | STRING |
| HIGH | Examples/torch/compression/spatial_svd.py | 0 | finetunes the model. the implemtation provided here is just an example, provide your own implementation if needed. :para | STRING |
| HIGH | Examples/torch/compression/channel_pruning.py | 0 | provides apis for model compression using aimet weight svd, evaluation and finetuning. | STRING |
| HIGH | Examples/torch/compression/weight_svd.py | 0 | provides apis for model compression using aimet weight svd, evaluation and finetuning. | STRING |
| HIGH | Examples/torch/compression/spatial_svd_cp.py | 0 | provides apis for model compression using aimet weight svd, evaluation and finetuning. | STRING |
| HIGH | Examples/torch/compression/spatial_svd.py | 0 | provides apis for model compression using aimet weight svd, evaluation and finetuning. | STRING |
| HIGH | Examples/torch/compression/channel_pruning.py | 0 | evaluate the specified model using the specified number of samples from the validation set. aimet's compress_model() exp | STRING |
| HIGH | Examples/torch/compression/weight_svd.py | 0 | evaluate the specified model using the specified number of samples from the validation set. aimet's compress_model() exp | STRING |
| HIGH | Examples/torch/compression/spatial_svd_cp.py | 0 | evaluate the specified model using the specified number of samples from the validation set. aimet's compress_model() exp | STRING |
| HIGH | Examples/torch/compression/spatial_svd.py | 0 | evaluate the specified model using the specified number of samples from the validation set. aimet's compress_model() exp | STRING |
| HIGH | Examples/torch/compression/channel_pruning.py | 0 | 1. instantiate data pipeline for evaluation and training 2. load the pretrained resnet18 model 3. calculate floating poi | STRING |
| HIGH | Examples/torch/compression/weight_svd.py | 0 | 1. instantiate data pipeline for evaluation and training 2. load the pretrained resnet18 model 3. calculate floating poi | STRING |
| HIGH | Examples/torch/compression/spatial_svd_cp.py | 0 | 1. instantiate data pipeline for evaluation and training 2. load the pretrained resnet18 model 3. calculate floating poi | STRING |
| HIGH | Examples/torch/compression/spatial_svd.py | 0 | 1. instantiate data pipeline for evaluation and training 2. load the pretrained resnet18 model 3. calculate floating poi | STRING |
| HIGH | NightlyTests/torch/test_compress_example_torch.py | 0 | constructor :param cuda: if true, data will be loaded in gpu memory :param seed: seed to use for randomization (to help | STRING |
| HIGH | NightlyTests/torch/models/mnist_torch_model.py | 0 | constructor :param cuda: if true, data will be loaded in gpu memory :param seed: seed to use for randomization (to help | STRING |
| HIGH | …tensions/torch/test/python/models/mnist_torch_model.py | 0 | constructor :param cuda: if true, data will be loaded in gpu memory :param seed: seed to use for randomization (to help | STRING |
| HIGH | …ions/torch/test/python/v2/models_/mnist_torch_model.py | 0 | constructor :param cuda: if true, data will be loaded in gpu memory :param seed: seed to use for randomization (to help | STRING |
| HIGH | …h/src/python/aimet_torch/examples/mnist_torch_model.py | 0 | constructor :param cuda: if true, data will be loaded in gpu memory :param seed: seed to use for randomization (to help | STRING |
| HIGH | NightlyTests/torch/models/mobilenet.py | 0 | loads model from a path :param model: :param model_path: :param location: :return: loaded model | STRING |
| HIGH | TrainingExtensions/onnx/test/python/models/mobilenet.py | 0 | loads model from a path :param model: :param model_path: :param location: :return: loaded model | STRING |
| HIGH | …ainingExtensions/torch/test/python/models/mobilenet.py | 0 | loads model from a path :param model: :param model_path: :param location: :return: loaded model | STRING |
| HIGH | …ons/torch/src/python/aimet_torch/examples/mobilenet.py | 0 | loads model from a path :param model: :param model_path: :param location: :return: loaded model | STRING |
| HIGH | NightlyTests/torch/models/mnist_torch_model.py | 0 | overriden implementation for the forward pass :param inputs: one or more inputs for the model :return: output of the for | STRING |
| HIGH | …tensions/torch/test/python/models/mnist_torch_model.py | 0 | overriden implementation for the forward pass :param inputs: one or more inputs for the model :return: output of the for | STRING |
| 303 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/plugins/local/aimet.py | 4 | CODE | |
| LOW | GenAILab/bench/analysis.py | 30 | CODE | |
| LOW | GenAILab/bench/yaml_config_parser.py | 6 | CODE | |
| LOW | GenAILab/bench/yaml_config_parser.py | 21 | CODE | |
| LOW | GenAILab/bench/recipe_cache.py | 27 | CODE | |
| LOW | GenAILab/bench/recipe_chain.py | 6 | CODE | |
| LOW | GenAILab/bench/precision.py | 6 | CODE | |
| LOW | GenAILab/bench/precision.py | 11 | CODE | |
| LOW | GenAILab/bench/precision.py | 22 | CODE | |
| LOW | GenAILab/bench/eval_context.py | 6 | CODE | |
| LOW | GenAILab/bench/grade_responses.py | 47 | CODE | |
| LOW | GenAILab/bench/onnx/analysis_passes.py | 6 | CODE | |
| LOW | GenAILab/bench/onnx/test_genai.py | 34 | CODE | |
| LOW | GenAILab/bench/onnx/test_genai.py | 35 | CODE | |
| LOW | GenAILab/bench/onnx/test_genai.py | 36 | CODE | |
| LOW | GenAILab/bench/onnx/test_genai.py | 37 | CODE | |
| LOW | GenAILab/bench/onnx/test_genai.py | 38 | CODE | |
| LOW | GenAILab/bench/torch/test_genai.py | 18 | CODE | |
| LOW | GenAILab/bench/torch/test_genai.py | 34 | CODE | |
| LOW | GenAILab/bench/torch/test_genai.py | 35 | CODE | |
| LOW | GenAILab/bench/utils/generation_utils.py | 10 | CODE | |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 6 | CODE | |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_utils.py | 8 | CODE | |
| LOW | GenAILab/tests/unit/onnx/test_generator_utils.py | 8 | CODE | |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_instantiation.py | 10 | CODE | |
| LOW | GenAILab/tests/unit/onnx/test_quantsim_instantiation.py | 12 | CODE | |
| LOW | GenAILab/tests/unit/torch/test_quantsim_utils.py | 6 | CODE | |
| LOW | GenAILab/tests/unit/torch/test_quantsim_utils.py | 8 | CODE | |
| LOW | GenAILab/tests/unit/torch/test_generator_utils.py | 8 | CODE | |
| LOW | …nAILab/tests/unit/torch/test_quantsim_instantiation.py | 12 | CODE | |
| LOW | …nAILab/tests/unit/torch/test_quantsim_instantiation.py | 15 | CODE | |
| LOW | …Lab/tests/unit/models/test_hub_compatible_generator.py | 7 | CODE | |
| LOW | …Lab/tests/unit/models/test_hub_compatible_generator.py | 8 | CODE | |
| LOW | GenAILab/tests/unit/models/test_generator.py | 6 | CODE | |
| LOW | GenAILab/tests/unit/models/test_vlm_generator.py | 11 | CODE | |
| LOW | GenAILab/tests/unit/models/test_layer_cache.py | 8 | CODE | |
| LOW | GenAILab/tests/unit/models/test_adaptations.py | 6 | CODE | |
| LOW | GenAILab/tests/unit/models/test_base.py | 8 | CODE | |
| LOW | GenAILab/tests/unit/infrastructure/test_main.py | 7 | CODE | |
| LOW | GenAILab/tests/unit/infrastructure/test_main.py | 10 | CODE | |
| LOW | GenAILab/tests/unit/helpers/test_model_cache.py | 6 | CODE | |
| LOW | GenAILab/tests/unit/helpers/test_yaml_config_parser.py | 6 | CODE | |
| LOW | GenAILab/tests/unit/helpers/test_yaml_config_parser.py | 7 | CODE | |
| LOW | GenAILab/tests/unit/helpers/test_yaml_config_parser.py | 12 | CODE | |
| LOW | GenAILab/tests/unit/helpers/test_precision_config.py | 8 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/__init__.py | 7 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/__init__.py | 10 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/__init__.py | 13 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/__init__.py | 16 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/__init__.py | 20 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/llm.py | 6 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/vlm.py | 6 | CODE | |
| LOW | …AILab/qai_hub_lm/backends/onnx/adaptations/__init__.py | 4 | CODE | |
| LOW | …Lab/qai_hub_lm/backends/onnx/adaptations/hub_models.py | 40 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/gemma4.py | 12 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/__init__.py | 9 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/__init__.py | 12 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/__init__.py | 15 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/__init__.py | 19 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/__init__.py | 27 | CODE | |
| 539 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | packaging/plugins/local/aimet.py | 186 | except Exception: | CODE |
| HIGH | GenAILab/bench/profiler.py | 172 | except ImportError: | CODE |
| HIGH | GenAILab/bench/profiler.py | 184 | except (subprocess.TimeoutExpired, FileNotFoundError): | CODE |
| HIGH | GenAILab/bench/model_cache.py | 128 | except Exception: | CODE |
| HIGH | GenAILab/bench/model_cache.py | 245 | except Exception: | CODE |
| HIGH | GenAILab/qai_hub_lm/backends/onnx/vlm.py | 468 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/backends/onnx/vlm.py | 478 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/backends/onnx/vlm.py | 488 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/backends/onnx/quantsim_utils.py | 79 | except AttributeError: | CODE |
| HIGH | GenAILab/qai_hub_lm/backends/torch/__init__.py | 28 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/backends/torch/vlm.py | 222 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/backends/torch/vlm.py | 232 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/models/utils/compat.py | 42 | except (ImportError, AttributeError): | CODE |
| HIGH | GenAILab/qai_hub_lm/models/utils/rope_embedding.py | 42 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/models/utils/rope_embedding.py | 51 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/models/utils/rope_embedding.py | 60 | except ImportError: | CODE |
| HIGH | GenAILab/qai_hub_lm/models/utils/rope_embedding.py | 69 | except ImportError: | CODE |
| HIGH | scripts/all/build_and_test.py | 349 | except Exception: | CODE |
| HIGH | scripts/all/build_and_test.py | 281 | except Exception: | CODE |
| HIGH | scripts/all/build_and_test.py | 316 | except Exception: | CODE |
| HIGH | scripts/all/util.py | 69 | except Exception: | CODE |
| HIGH | AIMETRegression/suite_runner.py | 352 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/onnx/_common.py | 277 | except (OSError, PermissionError): | CODE |
| HIGH | AIMETRegression/features/torch/quantsim_runner.py | 148 | except NotImplementedError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 111 | except (TypeError, ValueError): | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 167 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 175 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 183 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 191 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 199 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 214 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 224 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 245 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 254 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 264 | except (ImportError, ModuleNotFoundError): | CODE |
| HIGH | AIMETRegression/features/torch/_common.py | 300 | except (ImportError, ModuleNotFoundError): | CODE |
| HIGH | AIMETRegression/features/torch/utils.py | 153 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/utils.py | 192 | except ImportError: | CODE |
| HIGH | AIMETRegression/features/torch/utils.py | 144 | except Exception: | CODE |
| HIGH | AIMETRegression/features/torch/utils.py | 181 | except Exception: | CODE |
| HIGH | AIMETRegression/models/ai_hub_loader.py | 84 | except TypeError: | CODE |
| HIGH | AIMETRegression/workflow/utils.py | 102 | except: | CODE |
| HIGH | AIMETRegression/workflow/utils.py | 112 | except: | CODE |
| HIGH | AIMETRegression/workflow/utils.py | 140 | except: | CODE |
| HIGH | AIMETRegression/evaluation/eval_qnn.py | 52 | except Exception: | CODE |
| HIGH | AIMETRegression/evaluation/eval_qnn.py | 67 | except Exception: | CODE |
| HIGH | AIMETRegression/evaluation/eval_qnn.py | 236 | except Exception: | CODE |
| HIGH | AIMETRegression/evaluation/eval_qnn.py | 226 | except Exception: | CODE |
| HIGH | AIMETRegression/evaluation/eval_torch.py | 143 | except NotImplementedError: | CODE |
| HIGH | AIMETRegression/evaluation/metrics_utils.py | 128 | except Exception: | CODE |
| HIGH | AIMETRegression/evaluation/metrics_utils.py | 336 | except pynvml.NVMLError: | CODE |
| HIGH | AIMETRegression/evaluation/metrics_utils.py | 412 | except Exception: | CODE |
| HIGH | NightlyTests/onnx/test_quantsim.py | 21 | except (ImportError, OSError): | CODE |
| HIGH | NightlyTests/onnx/test_rnn_quantsim.py | 21 | except (ImportError, OSError): | CODE |
| HIGH | …ions/onnx/test/python/test_llm_topology_integration.py | 152 | except (AttributeError, TypeError): | CODE |
| HIGH | …/onnx/src/python/aimet_onnx/adascale/onnx2torch_ext.py | 43 | except (RuntimeError, defs.SchemaError): | CODE |
| HIGH | …/onnx/src/python/aimet_onnx/adascale/onnx2torch_ext.py | 158 | except KeyError: | CODE |
| HIGH | …/onnx/src/python/aimet_onnx/adascale/onnx2torch_ext.py | 216 | except KeyError: | CODE |
| HIGH | TrainingExtensions/torch/test/python/test_utils.py | 609 | except: | CODE |
| HIGH | TrainingExtensions/torch/test/python/test_utils.py | 653 | except: | CODE |
| 32 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | GenAILab/bench/torch/example_custom_script.py | 90 | # Create a generator object to accurately simulate inference with static graph constraints while maintaining the | COMMENT |
| MEDIUM | …nAILab/tests/unit/torch/test_quantsim_instantiation.py | 27 | # Create a module hierarchy that mimics a real model | COMMENT |
| MEDIUM | GenAILab/qai_hub_lm/backends/onnx/export_utils.py | 395 | # Create the checkpoint directory if it does not exist. | COMMENT |
| MEDIUM | Docs/snippets/onnx/apply_seqmse.py | 74 | # Create the QuantizationSimModel | COMMENT |
| MEDIUM | Docs/snippets/torch/apply_qwlora.py | 135 | # Create a QuantizationSimModel | COMMENT |
| MEDIUM | Examples/onnx/quantize.py | 220 | # Create a generator object to accurately simulate inference with static graph constraints while maintaining the | COMMENT |
| MEDIUM | Examples/torch/quantize.py | 215 | # Create a generator object to accurately simulate inference with static graph constraints while maintaining the | COMMENT |
| MEDIUM | Examples/torch/evaluate.py | 210 | # Create a generator object to accurately simulate inference with static graph constraints while maintaining the | COMMENT |
| MEDIUM | …ples/torch/quantization/qat_range_learning_ddp_eval.py | 179 | # Create an instance of DDP in order to perform distributed eval | COMMENT |
| MEDIUM | …tization/qat_range_learning_ddp_training_lightining.py | 217 | # Define the trainer here. | COMMENT |
| MEDIUM | Examples/torch/v2/quickstart_guide.py | 108 | # Create a model | COMMENT |
| MEDIUM | AIMETRegression/configs/_defaults.yaml | 3 | # This file defines default settings inherited by all tests. | COMMENT |
| MEDIUM | TrainingExtensions/onnx/test/python/test_quantsim.py | 1693 | # Create a new quantsim model | COMMENT |
| MEDIUM | TrainingExtensions/onnx/test/python/test_quantsim.py | 2140 | # Create a model with very high total activation memory usage | COMMENT |
| MEDIUM | TrainingExtensions/onnx/test/python/test_quantsim.py | 4307 | # Create an identical QuantizationSimModel | COMMENT |
| MEDIUM | …ingExtensions/onnx/test/python/test_mixed_precision.py | 498 | # Create an accuracy list | COMMENT |
| MEDIUM | …ingExtensions/onnx/test/python/test_mixed_precision.py | 592 | # Create an accuracy list | COMMENT |
| MEDIUM | …ningExtensions/onnx/test/python/test_qc_quantize_op.py | 1953 | # Create a new per-tensor quantizer | COMMENT |
| MEDIUM⚡ | …iningExtensions/onnx/test/python/test_libquant_info.py | 98 | # Create a BlockTensorQuantizer | COMMENT |
| MEDIUM | TrainingExtensions/onnx/test/python/test_adascale.py | 1327 | # Create a copy of the weights before applying AdaScale | COMMENT |
| MEDIUM | …Extensions/onnx/test/python/models/test_models_onnx.py | 50 | # Create the initializer tensor for the weights. | COMMENT |
| MEDIUM | …Extensions/onnx/test/python/models/test_models_onnx.py | 99 | # Create the initializer tensor for the weights. | COMMENT |
| MEDIUM | …Extensions/onnx/test/python/models/test_models_onnx.py | 158 | # Create the model (ModelProto) | COMMENT |
| MEDIUM | …Extensions/onnx/test/python/models/test_models_onnx.py | 195 | # Create the initializer tensor for the weights. | COMMENT |
| MEDIUM | …Extensions/onnx/test/python/models/test_models_onnx.py | 244 | # Create the initializer tensor for the weights. | COMMENT |
| MEDIUM | …Extensions/onnx/test/python/models/test_models_onnx.py | 307 | # Create the model (ModelProto) | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 2998 | # Define the fixed shape input tensor | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 3006 | # Define the constant 1D vector to be added to the input | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 3012 | # Define the weights for both matmuls | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 3020 | # Define the Add node | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 3028 | # Define the MatMul nodes | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 3043 | # Define the output tensors | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 3051 | # Create the graph | COMMENT |
| MEDIUM⚡ | …Extensions/onnx/test/python/models/models_for_tests.py | 3060 | # Create the model | COMMENT |
| MEDIUM | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 1360 | # Create a new initializer with unique name (use full module_path | COMMENT |
| MEDIUM | …ython/aimet_onnx/analysis/quant_stats_visualization.py | 757 | # Defining the default values of plotting parameters | STRING |
| MEDIUM⚡ | …ython/aimet_onnx/analysis/quant_stats_visualization.py | 789 | # Defining the table objects and name filter views | STRING |
| MEDIUM⚡ | …ython/aimet_onnx/analysis/quant_stats_visualization.py | 795 | # Defining a hover functionality to see layer details on hovering on the marker points and selections | STRING |
| MEDIUM⚡ | …ython/aimet_onnx/analysis/quant_stats_visualization.py | 800 | # Creating the input widgets | STRING |
| MEDIUM | …ython/aimet_onnx/analysis/quant_stats_visualization.py | 812 | # Define the formatting | STRING |
| MEDIUM | …onnx/src/python/aimet_onnx/sequential_mse/transform.py | 138 | # Create a new input tensor with symbolic dimensions and same name | COMMENT |
| MEDIUM | …x/src/python/aimet_onnx/adascale/adascale_optimizer.py | 502 | ] # Create a new tensor | CODE |
| MEDIUM | …x/src/python/aimet_onnx/adascale/adascale_optimizer.py | 571 | ] # Create a new tensor | CODE |
| MEDIUM | TrainingExtensions/torch/test/python/test_utils.py | 542 | # Create a model with a layer with multi-outputs | COMMENT |
| MEDIUM | …Extensions/torch/test/python/test_comp_ratio_select.py | 197 | # Create a layer database | COMMENT |
| MEDIUM | TrainingExtensions/torch/test/python/test_weight_svd.py | 219 | # Create a layer database | COMMENT |
| MEDIUM | …ainingExtensions/torch/test/python/test_spatial_svd.py | 152 | # Create a layer database | COMMENT |
| MEDIUM | …ainingExtensions/torch/test/python/test_spatial_svd.py | 198 | # Create a layer database | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_channel_pruning.py | 784 | # Create a layer database | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_channel_pruning.py | 831 | # Create a layer database | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_channel_pruning.py | 895 | # Create a layer database | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_channel_pruning.py | 952 | # Create a layer database | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_cost_calculator.py | 323 | # Create a list of tuples of (layer, comp_ratio) | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_cost_calculator.py | 348 | # Create a list of tuples of (layer, comp_ratio) | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_cost_calculator.py | 364 | # Create a list of tuples of (layer, comp_ratio) | COMMENT |
| MEDIUM | …ngExtensions/torch/test/python/test_cost_calculator.py | 435 | # Create a list of tuples of (layer, comp_ratio) | COMMENT |
| MEDIUM⚡ | …ngExtensions/torch/test/python/test_cost_calculator.py | 464 | # Create a list of tuples of (layer, comp_ratio) | COMMENT |
| MEDIUM⚡ | …ngExtensions/torch/test/python/test_cost_calculator.py | 474 | # Create the Input channel pruner | COMMENT |
| MEDIUM⚡ | …ions/torch/test/python/experimental/test_transforms.py | 50 | # Create a diagonal matrix with positive singular values | COMMENT |
| MEDIUM | …ions/torch/test/python/experimental/test_transforms.py | 172 | # Create a matrix transform op | COMMENT |
| 53 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/plugins/local/aimet.py | 87 | CODE | |
| LOW | GenAILab/__main__.py | 339 | CODE | |
| LOW | GenAILab/bench/metrics.py | 2486 | CODE | |
| LOW | GenAILab/bench/yaml_config_parser.py | 451 | CODE | |
| LOW | GenAILab/bench/yaml_config_parser.py | 562 | CODE | |
| LOW | GenAILab/bench/yaml_config_parser.py | 668 | CODE | |
| LOW | GenAILab/bench/recipe_cache.py | 76 | CODE | |
| LOW | GenAILab/bench/recipe_cache.py | 417 | CODE | |
| LOW | GenAILab/bench/recipe_cache.py | 446 | CODE | |
| LOW | GenAILab/bench/recipe_chain.py | 24 | CODE | |
| LOW | GenAILab/bench/precision.py | 243 | CODE | |
| LOW | GenAILab/bench/onnx/quant_recipes.py | 79 | CODE | |
| LOW | GenAILab/bench/torch/test_genai.py | 40 | CODE | |
| LOW | GenAILab/tests/unit/models/test_components.py | 276 | CODE | |
| LOW | GenAILab/tests/unit/schema/test_import_closure.py | 40 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/vlm.py | 66 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/quantsim_utils.py | 164 | CODE | |
| LOW | …AILab/qai_hub_lm/backends/onnx/torch_onnx_interface.py | 64 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/onnx/generator_utils.py | 147 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/gemma4.py | 163 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/gemma4.py | 185 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/vlm.py | 50 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/generator_utils.py | 57 | CODE | |
| LOW | GenAILab/qai_hub_lm/backends/torch/generator_utils.py | 42 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/qwen3_vl.py | 385 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/generator.py | 444 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/generator.py | 1145 | CODE | |
| LOW | GenAILab/qai_hub_lm/schema/precision.py | 188 | CODE | |
| LOW | GenAILab/qai_hub_lm/schema/recipe.py | 227 | CODE | |
| LOW | GenAILab/qai_hub_lm/schema/recipe.py | 253 | CODE | |
| LOW | GenAILab/qai_hub_lm/transforms/exportable_moe.py | 496 | CODE | |
| LOW | GenAILab/qai_hub_lm/transforms/exportable_moe.py | 503 | CODE | |
| LOW | scripts/all/build_and_test.py | 1240 | CODE | |
| LOW | scripts/all/build_and_test.py | 247 | CODE | |
| LOW | scripts/all/build_and_test.py | 356 | CODE | |
| LOW | AIMETRegression/suite_runner.py | 103 | CODE | |
| LOW | AIMETRegression/baseline_comparison.py | 996 | CODE | |
| LOW | AIMETRegression/baseline_comparison.py | 376 | CODE | |
| LOW | AIMETRegression/runner.py | 130 | CODE | |
| LOW | AIMETRegression/runner.py | 337 | CODE | |
| LOW | AIMETRegression/runner.py | 747 | CODE | |
| LOW | AIMETRegression/features/onnx/_common.py | 246 | CODE | |
| LOW | AIMETRegression/features/onnx/mixed_precision_runner.py | 67 | CODE | |
| LOW | …METRegression/features/torch/mixed_precision_runner.py | 103 | CODE | |
| LOW | AIMETRegression/features/torch/utils.py | 91 | CODE | |
| LOW | AIMETRegression/features/torch/utils.py | 158 | CODE | |
| LOW | AIMETRegression/workflow/summary.py | 83 | CODE | |
| LOW | AIMETRegression/workflow/utils.py | 87 | CODE | |
| LOW | AIMETRegression/workflow/utils.py | 166 | CODE | |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 171 | CODE | |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 246 | CODE | |
| LOW | AIMETRegression/evaluation/eval_torch.py | 64 | CODE | |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 216 | CODE | |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 319 | CODE | |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 217 | CODE | |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 348 | CODE | |
| LOW | NightlyTests/onnx/test_mixed_precision.py | 157 | CODE | |
| LOW | NightlyTests/torch/dependencies.py | 120 | CODE | |
| LOW | NightlyTests/torch/models/mobilenet.py | 206 | CODE | |
| LOW | NightlyTests/torch/models/mobilenet.py | 279 | CODE | |
| 345 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/INSTALL.txt | 1 | # ================================================================================== | COMMENT |
| LOW | packaging/NOTICE.txt | 61 | COMMENT | |
| LOW | packaging/NOTICE.txt | 81 | // distribution. | COMMENT |
| LOW | packaging/NOTICE.txt | 101 | # | COMMENT |
| LOW | packaging/NOTICE.txt | 121 | # SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT | COMMENT |
| LOW | packaging/NOTICE.txt | 141 | // in the documentation and/or other materials provided with the | COMMENT |
| LOW | packaging/NOTICE.txt | 161 | // Copyright 2008, Google Inc. | COMMENT |
| LOW | packaging/NOTICE.txt | 181 | // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT | COMMENT |
| LOW | packaging/NOTICE.txt | 201 | // * Redistributions in binary form must reproduce the above | COMMENT |
| LOW | packaging/NOTICE.txt | 221 | ================================================================================= | COMMENT |
| LOW | packaging/NOTICE.txt | 281 | // | COMMENT |
| LOW | packaging/NOTICE.txt | 301 | // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT | COMMENT |
| LOW | packaging/convert_wheel_format.sh | 1 | #!/bin/bash | COMMENT |
| LOW | …ions/DlQuantization/test/TestQuantizationFunctions.cpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …izations/DlQuantization/test/test_quantization_lib.hpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …ions/DlQuantization/src/PercentileEncodingAnalyzer.cpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …delOptimizations/DlQuantization/src/math_functions.cpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …delOptimizations/DlQuantization/src/trim_functions.hpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …delOptimizations/DlQuantization/src/trim_functions.cpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …elOptimizations/DlQuantization/src/TensorQuantizer.cpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …izations/DlQuantization/src/MinMaxEncodingAnalyzer.cpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …ptimizations/DlQuantization/src/quantization_utils.cpp | 1 | // Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/bench/yaml_config_parser.py | 701 | # | COMMENT |
| LOW | GenAILab/bench/onnx/__init__.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/bench/onnx/quant_recipes.py | 321 | num_batches: int = 32, | COMMENT |
| LOW | GenAILab/bench/onnx/test_genai.py | 121 | audio_frames=audio_frames, | COMMENT |
| LOW | GenAILab/bench/utils/__init__.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/bench/prompts/calibration_prompts.yaml | 1 | # Hand-curated seed prompts for SELF-GENERATED CALIBRATION data. | COMMENT |
| LOW | GenAILab/bench/prompts/multimodal_prompts.yaml | 1 | # Prompts for evaluating quantized VLM generation quality. | COMMENT |
| LOW | GenAILab/tests/__init__.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/tests/integration/__init__.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/qai_hub_lm/__init__.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/qai_hub_lm/models/__init__.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/qai_hub_lm/models/qwen3_5.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/qai_hub_lm/models/components.py | 61 | # --- input-encoder only ------------------------------------------------- | COMMENT |
| LOW | GenAILab/qai_hub_lm/models/utils/compat.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | GenAILab/qai_hub_lm/models/utils/__init__.py | 1 | # Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 201 | g = g * mask | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 601 | COMMENT | |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 621 | # * Run them AFTER quantsim export. They delete and rewrite nodes; running | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 641 | # Everything else depends on this producing genuinely static shapes. | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 661 | # 3. Re-run shape inference, then constant-fold (pass 1). | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 681 | # consume. They remove no arithmetic (the MAC count is unchanged; the ops are | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 701 | # and ~14 elementwise ops per layer computing nothing, for 44 MACs of real | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 721 | # | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 741 | # cannot reach the Mul. State that precondition explicitly in the pass -- it is | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 761 | # | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 781 | # for free from a backend. | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 801 | # Appendix A -- The conv roll is NOT a graph rewrite | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 821 | # hardened, both specializations : 625 nodes | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 841 | # =========================================================================== | COMMENT |
| LOW | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 861 | # on a small fixture graph. They are specified here only because this is where | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 221 | # each running whichever expert is selected (``ElementWiseMux`` over the | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 241 | # with a ``force_all`` switch and an input mask. passes 4-5 | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 261 | # ``pkg.torch.onnx.name_scopes`` metadata (128/128 on the dense block), so | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 281 | # P0 The subselect chain (ConstantOfShape, ScatterElements, masking Mul) | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 301 | # per slot is open (A2) -- no pass can fix that. | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 321 | # ``ORT_ENABLE_BASIC``, with the same warning as linear attention: not | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 341 | # WHAT. Inside the ``gate`` scope, the chain is the Softmax output's ReduceMax | COMMENT |
| LOW | GenAILab/qai_hub_lm/transforms/expert_subselection.py | 361 | # For Qwen3-/Qwen3.5-MoE the indices come from a TopK over the expert axis, so | COMMENT |
| 111 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/plugins/local/aimet.py | 61 | # Check if ARM64 is specified in CMAKE_ARGS or if running natively on ARM64 | COMMENT |
| LOW | GenAILab/bench/profiler.py | 437 | # Check if the file exists | COMMENT |
| LOW | GenAILab/bench/profiler.py | 439 | # Open the file and load the existing data | COMMENT |
| LOW | Docs/legacy/torch_code_examples/mixed_precision.py | 87 | # Set clean_start to False to start from an existing cache | COMMENT |
| LOW | Docs/legacy/torch_code_examples/mixed_precision.py | 88 | # Set allowed_accuracy_drop to 0.9 to export the 90% drop point in pareto list | COMMENT |
| LOW | scripts/posix/install_macos_deps.sh | 18 | # Check if Homebrew is installed | COMMENT |
| LOW⚡ | scripts/all/build_and_test.py | 357 | # Check if we can skip the build | COMMENT |
| LOW | scripts/all/build_and_test.py | 617 | # Check if we can skip wheel installation | COMMENT |
| LOW | scripts/kube/dev.sh | 85 | # Check if it's a pod name or a workflow name | COMMENT |
| LOW | scripts/kube/launch_pod.sh | 109 | # Check if workflow has failed | COMMENT |
| LOW | .github/workflows/dco.yml | 34 | # Check if author is indeed signing off commit | COMMENT |
| LOW | AIMETRegression/features/onnx/lite_mp_runner.py | 89 | # Check if already a valid symbol | COMMENT |
| LOW | …METRegression/features/torch/mixed_precision_runner.py | 202 | # Check if user requested unsupported candidates | COMMENT |
| LOW | AIMETRegression/features/torch/utils.py | 110 | # Check if already patched by another mechanism | COMMENT |
| LOW | AIMETRegression/models/ai_hub_loader.py | 77 | # Check if it's a class and subclass of BaseModel | COMMENT |
| LOW | NightlyTests/torch/dependencies.py | 35 | # Check if we need to generate the .pth for CPU or GPU. If not, return | COMMENT |
| LOW⚡ | TrainingExtensions/onnx/test/python/test_quantsim.py | 290 | # Check if qc quantize op node is correctly connect to the corresponding onnx node | COMMENT |
| LOW⚡ | TrainingExtensions/onnx/test/python/test_quantsim.py | 297 | # Check if op_mode is set correctly for each qc quantize op node | COMMENT |
| LOW | …ingExtensions/onnx/test/python/test_mixed_precision.py | 253 | # Check if quantizer group is set to maximum bitwidth | COMMENT |
| LOW | …ingExtensions/onnx/test/python/test_mixed_precision.py | 320 | # Check if accuracy list is in descending order | COMMENT |
| LOW | …ningExtensions/onnx/test/python/test_quant_analyzer.py | 194 | # Check if it is exported to correct html file. | COMMENT |
| LOW | …ningExtensions/onnx/test/python/test_quant_analyzer.py | 231 | # Check if it is exported to correct html file. | COMMENT |
| LOW | …ningExtensions/onnx/test/python/test_quant_analyzer.py | 438 | # Check if it is exported to correct html file. | COMMENT |
| LOW⚡ | …ons/onnx/test/python/test_graph_passes/test_rmsnorm.py | 43 | # Check if quantization is disabled for RMSNormalization intermediate op outputs | COMMENT |
| LOW⚡ | …ons/onnx/test/python/test_graph_passes/test_rmsnorm.py | 45 | # Check if quantization is enabled for last op of RMSNormalization sub-graph | COMMENT |
| LOW⚡ | …ons/onnx/test/python/test_graph_passes/test_rmsnorm.py | 48 | # Check if quantization is disabled for RMSNormalization sub-graph constant ops except weight | COMMENT |
| LOW⚡ | …s/onnx/test/python/test_graph_passes/test_layernorm.py | 35 | # Check if quantization is disabled for LayerNormalization intermediate op outputs | COMMENT |
| LOW⚡ | …s/onnx/test/python/test_graph_passes/test_layernorm.py | 37 | # Check if quantization is enabled for last op of LayerNormalization sub-graph | COMMENT |
| LOW⚡ | …s/onnx/test/python/test_graph_passes/test_layernorm.py | 40 | # Check if quantization is disabled for LayerNormalization sub-graph constant ops except layernorm.weight | COMMENT |
| LOW⚡ | …s/onnx/test/python/test_graph_passes/test_layernorm.py | 68 | # Check if quantization is disabled for LayerNormalization intermediate op outputs | COMMENT |
| LOW⚡ | …s/onnx/test/python/test_graph_passes/test_layernorm.py | 70 | # Check if quantization is enabled for last op of LayerNormalization sub-graph | COMMENT |
| LOW⚡ | …s/onnx/test/python/test_graph_passes/test_layernorm.py | 72 | # Check if quantization is disabled for LayerNormalization sub-graph constant ops except layernorm.weight | COMMENT |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 4616 | # Check if encoding corresponds to non-quantizable tensor type | COMMENT |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 886 | # Check if the tensor is param to certain ops (eg: Resize) | STRING |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 906 | # Check if the tensor is output of certain ops | STRING |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 1196 | # Check if weights are constant | STRING |
| LOW | TrainingExtensions/onnx/src/python/aimet_onnx/utils.py | 177 | # Check if other node takes input from removed node | COMMENT |
| LOW | TrainingExtensions/onnx/src/python/aimet_onnx/utils.py | 434 | # Check if the weight is transposed before entering the node | COMMENT |
| LOW | …tensions/onnx/src/python/aimet_onnx/batch_norm_fold.py | 220 | # Check if this is actually a fully connected layer or a dynamic matmul | COMMENT |
| LOW | …x/src/python/aimet_onnx/adaround/adaround_optimizer.py | 97 | # Check if we can cache intermediate activation data. | COMMENT |
| LOW | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 431 | # Check if input comes from an Identity whose input is an initializer | COMMENT |
| LOW | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 471 | # Check if this Mul's inputs include a scale (Constant or init) | COMMENT |
| LOW | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 1503 | # Check if scale is already a named initializer | COMMENT |
| LOW | …rc/python/aimet_onnx/experimental/lora/peft_to_onnx.py | 806 | # Check if this input comes from a Constant node | COMMENT |
| LOW | …/python/aimet_onnx/experimental/lora/lora_configure.py | 632 | # Check if this input comes from a Constant node | COMMENT |
| LOW | …x/src/python/aimet_onnx/graph_passes/passes/rmsnorm.py | 68 | # Check if weights are present, skipping any Cast ops (e.g. from | COMMENT |
| LOW | …src/python/aimet_onnx/graph_passes/passes/layernorm.py | 93 | # Check if affine_transform is set. | COMMENT |
| LOW | …nnx/graph_passes/cleanup/remove_duplicate_qdq_pairs.py | 121 | # Check if Q1/DQ1 params match Q2/DQ2 params | COMMENT |
| LOW | …onnx/src/python/aimet_onnx/sequential_mse/transform.py | 475 | # Check if this is a Transpose op | COMMENT |
| LOW | …onnx/src/python/aimet_onnx/sequential_mse/transform.py | 491 | # Check if input is a leaf initializer (unquantized case) | COMMENT |
| LOW | …xtensions/torch/test/python/test_layer_output_utils.py | 168 | # Check if every layer-name has corresponding onnx layer-output-name | COMMENT |
| LOW⚡ | …tensions/torch/test/python/v1/test_quantizer_groups.py | 39 | # Check if there is one parent and one child | COMMENT |
| LOW | …/torch/test/python/experimental/test_let_quantizers.py | 244 | # Set scale to 2 | COMMENT |
| LOW | …sions/torch/test/python/experimental/test_omniquant.py | 342 | # Set model to lora model | COMMENT |
| LOW⚡ | …xtensions/torch/test/python/v2/test_mixed_precision.py | 397 | # Check if accuracy list is in descending order | COMMENT |
| LOW⚡ | …xtensions/torch/test/python/v2/test_mixed_precision.py | 435 | # Check if parsed list is sorted | COMMENT |
| LOW | …xtensions/torch/test/python/v2/test_mixed_precision.py | 287 | # Check if quantizer group is set to maximum bitwidth | COMMENT |
| LOW | …xtensions/torch/test/python/v2/test_mixed_precision.py | 327 | # Check if accuracy list is in descending order | COMMENT |
| LOW | …xtensions/torch/test/python/v2/test_mixed_precision.py | 357 | # Check if quantizer group is set to maximum bitwidth | COMMENT |
| LOW | …xtensions/torch/test/python/v2/test_mixed_precision.py | 476 | # Check if parsed list is sorted | COMMENT |
| 21 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 1570 | self.model.model.graph.node.extend(sim_nodes) | STRING |
| CRITICAL⚡ | …ensions/torch/test/python/v2/ab_test/test_quantizer.py | 1036 | grad_single_gpu = sim.model.conv1.weight.grad.clone().detach() | CODE |
| CRITICAL⚡ | …ensions/torch/test/python/v2/ab_test/test_quantizer.py | 1041 | weight_multi_gpu = sim.model.module.conv1.weight.clone().detach() | CODE |
| CRITICAL⚡ | …ensions/torch/test/python/v2/ab_test/test_quantizer.py | 1045 | grad_multi_gpu = sim.model.module.conv1.weight.grad.clone().detach() | CODE |
| CRITICAL⚡ | …ensions/torch/test/python/v2/ab_test/test_quantizer.py | 2994 | grad_single_gpu = sim.model.conv1.weight.grad.clone().detach() | CODE |
| CRITICAL⚡ | …ensions/torch/test/python/v2/ab_test/test_quantizer.py | 2999 | weight_multi_gpu = sim.model.module.conv1.weight.clone().detach() | CODE |
| CRITICAL⚡ | …ensions/torch/test/python/v2/ab_test/test_quantizer.py | 3003 | grad_multi_gpu = sim.model.module.conv1.weight.grad.clone().detach() | CODE |
| CRITICAL | …rc/python/aimet_torch/spinquant/spinquant_optimizer.py | 759 | visual_encoder.patch_embed.proj.weight.data.clone() | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH⚡ | GenAILab/bench/metrics.py | 609 | """KL divergence KL(P_fp || P_quant) over answer choice distributions.""" | STRING |
| HIGH⚡ | GenAILab/bench/metrics.py | 619 | """Reverse KL divergence KL(P_quant || P_fp) over answer choice distributions.""" | STRING |
| HIGH | GenAILab/bench/metrics.py | 653 | """KL divergence KL(P_fp || P_quant) over MMLU answer choice distributions.""" | STRING |
| HIGH | GenAILab/bench/metrics.py | 658 | """Reverse KL divergence KL(P_quant || P_fp) over MMLU answer choice distributions.""" | STRING |
| HIGH | GenAILab/bench/metrics.py | 866 | """KL divergence KL(P_fp || P_quant) over MMMU answer choice distributions.""" | STRING |
| HIGH | GenAILab/bench/metrics.py | 871 | """Reverse KL divergence KL(P_quant || P_fp) over MMMU answer choice distributions.""" | STRING |
| HIGH | GenAILab/bench/metrics.py | 1048 | """KL divergence KL(P_fp || P_quant) over ERQA answer choice distributions.""" | STRING |
| HIGH | GenAILab/bench/metrics.py | 1053 | """Reverse KL divergence KL(P_quant || P_fp) over ERQA answer choice distributions.""" | STRING |
| HIGH | …mples/torch/quantization/zero_bit_quantization_demo.py | 56 | - Compression ratio: infinity (original_size / 0 = undefined, but trust us) | STRING |
| HIGH | …ython/aimet_onnx/analysis/quant_stats_visualization.py | 522 | if (Math.abs(value) < 1e-3 || Math.abs(value) > 1e5) { | CODE |
| HIGH | …ython/aimet_onnx/analysis/quant_stats_visualization.py | 550 | if (Math.abs(value) < 1e-3 || Math.abs(value) > 1e5) { | CODE |
| HIGH | …h/src/python/aimet_torch/quantization/affine/_utils.py | 69 | undefined, & \text{otherwise} | CODE |
| HIGH | …h/src/python/aimet_torch/quantization/affine/_utils.py | 130 | undefined, & \text{otherwise} | CODE |
| HIGH | …torch/visualization_tools/quant_stats_visualization.py | 843 | if (Math.abs(value) < 1e-3 || Math.abs(value) > 1e5) { | CODE |
| HIGH | …torch/visualization_tools/quant_stats_visualization.py | 871 | if (Math.abs(value) < 1e-3 || Math.abs(value) > 1e5) { | CODE |
| HIGH | …ningExtensions/common/src/python/_aimet_common/defs.py | 360 | param_dtype: QuantizationDataType = QuantizationDataType.undefined, | CODE |
| HIGH | …ensions/common/src/python/_aimet_common/winnow/mask.py | 750 | :param is_null_connectivity: True if op is null connectivity, False otherwise | STRING |
| HIGH⚡ | …common/src/python/_aimet_common/winnow/winnow_utils.py | 58 | "Conv": ConnectivityType.null, | CODE |
| HIGH⚡ | …common/src/python/_aimet_common/winnow/winnow_utils.py | 59 | "ConvTranspose": ConnectivityType.null, | CODE |
| HIGH⚡ | …common/src/python/_aimet_common/winnow/winnow_utils.py | 60 | "Linear": ConnectivityType.null, | CODE |
| HIGH⚡ | …common/src/python/_aimet_common/winnow/winnow_utils.py | 61 | "Gemm": ConnectivityType.null, | CODE |
| HIGH | …common/src/python/_aimet_common/winnow/winnow_utils.py | 91 | "index_select": ConnectivityType.null, | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | AIMETRegression/suite_runner.py | 54 | Load and validate a suite configuration file. Suite files define: - profile: Which runtime profile to use | STRING |
| HIGH | AIMETRegression/config_loader.py | 42 | Load and merge configuration from multiple sources. This is the main entry point for config loading. It merges | STRING |
| HIGH | AIMETRegression/config_loader.py | 171 | List all test names defined in a model configuration. This is useful for: - Discovering what tests exist i | STRING |
| HIGH | AIMETRegression/features/onnx/quantsim_runner.py | 53 | Apply AIMET QuantSim to simulate INT8 quantization on an FP32 ONNX model. QuantSim creates a simulation model | STRING |
| HIGH | AIMETRegression/features/onnx/_common.py | 59 | Select available ORT execution providers from a preference list. This function ensures explicit provider selec | STRING |
| HIGH | AIMETRegression/features/onnx/_common.py | 154 | Construct an AIMET QuantizationSimModel from an ONNX file. CRITICAL DESIGN DECISION: This function loads t | STRING |
| HIGH | AIMETRegression/features/onnx/adaround_runner.py | 137 | Apply AdaRound optimization for improved weight quantization. AdaRound optimizes the rounding of weights durin | STRING |
| HIGH | AIMETRegression/features/onnx/mixed_precision_runner.py | 153 | Apply automatic mixed-precision optimization using AIMET's search algorithm. This technique uses AIMET's choos | STRING |
| HIGH | AIMETRegression/features/onnx/lite_mp_runner.py | 70 | Convert configuration value to AIMET's required precision symbol. AIMET's flip_layers_to_higher_precision API | STRING |
| HIGH | AIMETRegression/features/onnx/lite_mp_runner.py | 129 | Apply Lite Mixed-Precision quantization to an FP32 ONNX model. This technique provides an optimal accuracy-per | STRING |
| HIGH | AIMETRegression/features/torch/_common.py | 72 | Map configuration string to AIMET QuantScheme enum. Args: scheme_str: Scheme name from config (e.g., " | STRING |
| HIGH | AIMETRegression/features/torch/adaround_runner.py | 71 | Apply AdaRound optimization for improved weight quantization on PyTorch models. AdaRound optimizes the roundin | STRING |
| HIGH | AIMETRegression/models/ai_hub_loader.py | 39 | Find and return the appropriate model class from a module. QAI Hub Models modules may contain the model class | STRING |
| HIGH | AIMETRegression/models/ai_hub_loader.py | 133 | Load a model from QAI Hub Models with its associated dataset and specifications. This is the main entry point | STRING |
| HIGH | AIMETRegression/evaluation/metrics_utils.py | 71 | Format duration with automatic unit selection for readability. This function chooses the most appropriate form | STRING |
| HIGH | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 3202 | Returns graph path from a tensor to the "effective" quantizer associated with it (if any). Node Q is t | STRING |
| HIGH | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 4200 | Convert an ONNX model and its AIMET encodings into an ONNX QDQ model. This is a direct translation: the encodi | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/plugins/local/aimet.py | 186 | except Exception: | CODE |
| LOW⚡ | GenAILab/bench/yaml_config_parser.py | 535 | except Exception as e: | CODE |
| LOW | GenAILab/bench/model_cache.py | 128 | except Exception: | CODE |
| LOW | GenAILab/bench/model_cache.py | 245 | except Exception: | CODE |
| LOW⚡ | GenAILab/qai_hub_lm/backends/onnx/export_utils.py | 40 | except Exception: | CODE |
| LOW⚡ | scripts/all/build_and_test.py | 349 | except Exception: | CODE |
| LOW⚡ | scripts/all/build_and_test.py | 351 | except Exception as e: | CODE |
| LOW | scripts/all/build_and_test.py | 281 | except Exception: | CODE |
| LOW | scripts/all/build_and_test.py | 283 | except Exception as e: | CODE |
| LOW | scripts/all/build_and_test.py | 316 | except Exception: | CODE |
| LOW | scripts/all/build_and_test.py | 318 | except Exception as e: | CODE |
| LOW | scripts/all/build_and_test.py | 1334 | except Exception as e: | CODE |
| LOW | scripts/all/util.py | 47 | except Exception: | CODE |
| LOW | scripts/all/util.py | 69 | except Exception: | CODE |
| LOW | scripts/all/util.py | 101 | except Exception: | CODE |
| LOW | scripts/all/util.py | 117 | except Exception: | CODE |
| LOW | AIMETRegression/suite_runner.py | 269 | except Exception as e: | STRING |
| LOW | AIMETRegression/suite_runner.py | 276 | except Exception as e: | STRING |
| LOW | AIMETRegression/suite_runner.py | 374 | except Exception as e: | STRING |
| LOW | AIMETRegression/runner.py | 108 | except Exception as e: | CODE |
| LOW | AIMETRegression/runner.py | 191 | except Exception as e: | CODE |
| LOW | AIMETRegression/runner.py | 211 | except Exception as e2: | CODE |
| LOW | AIMETRegression/runner.py | 227 | except Exception as e: | CODE |
| LOW | AIMETRegression/runner.py | 568 | except Exception as e: | CODE |
| LOW | AIMETRegression/runner.py | 645 | except Exception as e: | CODE |
| LOW | AIMETRegression/runner.py | 678 | except Exception as e: | CODE |
| LOW | AIMETRegression/runner.py | 814 | except Exception as e: | CODE |
| LOW | AIMETRegression/features/torch/_common.py | 277 | except Exception as error: # pylint: disable=broad-except | CODE |
| LOW | AIMETRegression/features/torch/_common.py | 291 | except Exception as error: # pylint: disable=broad-except | CODE |
| LOW | …METRegression/features/torch/mixed_precision_runner.py | 152 | except Exception as e: | CODE |
| LOW | …METRegression/features/torch/mixed_precision_runner.py | 317 | except Exception as e: | CODE |
| LOW | AIMETRegression/features/torch/utils.py | 144 | except Exception: | CODE |
| LOW | AIMETRegression/features/torch/utils.py | 181 | except Exception: | CODE |
| LOW | AIMETRegression/workflow/utils.py | 228 | except Exception as e: | STRING |
| LOW | AIMETRegression/workflow/artifacts.py | 528 | except Exception as e: | CODE |
| LOW | AIMETRegression/workflow/artifacts.py | 592 | except Exception as e: | CODE |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 52 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 67 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 88 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 226 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 236 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 287 | except Exception as e: | CODE |
| LOW | AIMETRegression/evaluation/eval_qnn.py | 299 | except Exception as e: | CODE |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 128 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 159 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 166 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 249 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 339 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 381 | except Exception: | CODE |
| LOW | AIMETRegression/evaluation/metrics_utils.py | 412 | except Exception: | CODE |
| LOW | AIMETRegression/report/report_writer.py | 114 | except Exception: | CODE |
| LOW | TrainingExtensions/onnx/test/python/test_lora.py | 1788 | except Exception: | CODE |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 4148 | except Exception as e: | CODE |
| LOW | TrainingExtensions/onnx/src/python/aimet_onnx/utils.py | 666 | except Exception as e: # pylint: disable=broad-except | CODE |
| LOW | TrainingExtensions/onnx/src/python/aimet_onnx/utils.py | 757 | except Exception as e: # pylint: disable=broad-except | CODE |
| LOW | …onnx/src/python/aimet_onnx/adascale/model_converter.py | 41 | except Exception: | CODE |
| LOW | …nsions/torch/src/python/aimet_torch/deepspeed_utils.py | 96 | except Exception as e: | CODE |
| LOW | TrainingExtensions/torch/src/python/aimet_torch/onnx.py | 191 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | …ainingExtensions/torch/src/python/aimet_torch/utils.py | 1260 | except Exception: | CODE |
| LOW | …ions/torch/src/python/aimet_torch/blockwise_sampler.py | 88 | except Exception as e: | CODE |
| 10 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 285 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 288 | # Step 2: Pre-compute param encodings with a real input so that | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 295 | # Step 3: Optimize weight quantization parameters to minimize layer-wise MSE. | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 326 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 329 | # Step 2: Pre-compute param encodings with a real input (same | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 333 | # Step 3: Optimize quantization parameters using AdaScale. The decoder-stack | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 246 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 251 | # Step 2: Optimize weight quantization parameters to minimize layer-wise MSE. | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 273 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 278 | # Step 2: Optimize quantization parameters using AdaScale. | COMMENT |
| LOW | scripts/README.md | 76 | # Step 1: Set up environment without installing AIMET | COMMENT |
| LOW | scripts/README.md | 79 | # Step 2: Build from source (interactive — prompts for variant selection) | COMMENT |
| LOW | AIMETRegression/config_loader.py | 82 | # Step 1: Load defaults (foundation for all configs) | COMMENT |
| LOW | AIMETRegression/config_loader.py | 96 | # Step 2: Merge profile (if specified) | COMMENT |
| LOW | AIMETRegression/config_loader.py | 119 | # Step 3: Load model configuration | COMMENT |
| LOW | AIMETRegression/config_loader.py | 150 | # Step 4: Find and merge specific test configuration | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 108 | # Step 1: Find LoRA patterns via topology-based graph traversal | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 116 | # Step 2: Validate against adapter config target_modules | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 137 | # Step 3: Split shared initializers (dynamo=False shares zero-init tensors) | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 140 | # Step 4: Rename inits to PEFT convention + insert Transpose for shape compat | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 176 | # Step 5: Convert scale Constants to named initializers | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 179 | # Step 6: Trace LoRA activations | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 182 | # Step 7: Build lora_names structure | COMMENT |
| LOW⚡ | …et_onnx/experimental/lora/lora_adapter_quantization.py | 946 | # Step 3: Set quantizer encodings | COMMENT |
| LOW⚡ | …et_onnx/experimental/lora/lora_adapter_quantization.py | 949 | # Step 4: Set bitwidths | COMMENT |
| LOW⚡ | …et_onnx/experimental/lora/lora_adapter_quantization.py | 952 | # Step 5: Rebuild session to pick up new initializer data | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 514 | # Step 1: Multi-adapter name insertion (lora_A.weight → lora_A.{adapter}.weight) | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 528 | # Step 2: If ONNX param names provided, match by suffix for any | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 573 | # Step 3: Transpose PEFT-convention weights to MatMul convention | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 904 | # Step 1: Remap safetensors keys → ONNX init names (with transposition) | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 915 | # Step 2: Update initializer data | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 163 | # Step 1: Identify the "Add" nodes where the 1st adapter branches merge back into the main graph | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 177 | # Step 2: Identify all LoRA adapter chains | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 188 | # Step 3: Create a dictionary of Add nodes in the chain to the full chain for quick lookup | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 196 | # Step 4: Create a map of attach point to the corresponding adapter chains | COMMENT |
| LOW | …ensions/torch/test/python/v2/quantsim/test_quantsim.py | 1277 | # NOTE: This module will ALWAYS fail when forward is called, | COMMENT |
| LOW⚡ | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 715 | # # Step 1: get a list of conversion costs in phase 2 solution | COMMENT |
| LOW⚡ | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 724 | # Step 2: sort the conversion costs list | COMMENT |
| LOW⚡ | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 728 | # Step 3: compute cumulative sums for the sorted conversion costs and determine the threshold | COMMENT |
| LOW | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 739 | # Step 4: Enforce the constraint | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 285 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 288 | # Step 2: Pre-compute param encodings with a real input so that | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 295 | # Step 3: Optimize weight quantization parameters to minimize layer-wise MSE. | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 326 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 329 | # Step 2: Pre-compute param encodings with a real input (same | COMMENT |
| LOW⚡ | GenAILab/bench/onnx/quant_recipes.py | 333 | # Step 3: Optimize quantization parameters using AdaScale. The decoder-stack | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 246 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 251 | # Step 2: Optimize weight quantization parameters to minimize layer-wise MSE. | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 273 | # Step 1: Collect calibration inputs in FP mode. | COMMENT |
| LOW⚡ | GenAILab/bench/torch/quant_recipes.py | 278 | # Step 2: Optimize quantization parameters using AdaScale. | COMMENT |
| LOW | AIMETRegression/config_loader.py | 82 | # Step 1: Load defaults (foundation for all configs) | COMMENT |
| LOW | AIMETRegression/config_loader.py | 96 | # Step 2: Merge profile (if specified) | COMMENT |
| LOW | AIMETRegression/config_loader.py | 119 | # Step 3: Load model configuration | COMMENT |
| LOW | AIMETRegression/config_loader.py | 150 | # Step 4: Find and merge specific test configuration | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 108 | # Step 1: Find LoRA patterns via topology-based graph traversal | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 116 | # Step 2: Validate against adapter config target_modules | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 137 | # Step 3: Split shared initializers (dynamo=False shares zero-init tensors) | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 140 | # Step 4: Rename inits to PEFT convention + insert Transpose for shape compat | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 176 | # Step 5: Convert scale Constants to named initializers | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 179 | # Step 6: Trace LoRA activations | COMMENT |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 182 | # Step 7: Build lora_names structure | COMMENT |
| LOW⚡ | …et_onnx/experimental/lora/lora_adapter_quantization.py | 946 | # Step 3: Set quantizer encodings | COMMENT |
| LOW⚡ | …et_onnx/experimental/lora/lora_adapter_quantization.py | 949 | # Step 4: Set bitwidths | COMMENT |
| LOW⚡ | …et_onnx/experimental/lora/lora_adapter_quantization.py | 952 | # Step 5: Rebuild session to pick up new initializer data | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 514 | # Step 1: Multi-adapter name insertion (lora_A.weight → lora_A.{adapter}.weight) | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 528 | # Step 2: If ONNX param names provided, match by suffix for any | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 573 | # Step 3: Transpose PEFT-convention weights to MatMul convention | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 904 | # Step 1: Remap safetensors keys → ONNX init names (with transposition) | COMMENT |
| LOW | …et_onnx/experimental/lora/lora_adapter_quantization.py | 915 | # Step 2: Update initializer data | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 163 | # Step 1: Identify the "Add" nodes where the 1st adapter branches merge back into the main graph | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 177 | # Step 2: Identify all LoRA adapter chains | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 188 | # Step 3: Create a dictionary of Add nodes in the chain to the full chain for quick lookup | COMMENT |
| LOW⚡ | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 196 | # Step 4: Create a map of attach point to the corresponding adapter chains | COMMENT |
| LOW⚡ | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 715 | # # Step 1: get a list of conversion costs in phase 2 solution | COMMENT |
| LOW⚡ | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 724 | # Step 2: sort the conversion costs list | COMMENT |
| LOW⚡ | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 728 | # Step 3: compute cumulative sums for the sorted conversion costs and determine the threshold | COMMENT |
| LOW | …/src/python/_aimet_common/amp/convert_ops_reduction.py | 739 | # Step 4: Enforce the constraint | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/plugins/local/aimet.py | 14 | __all__ = ["dynamic_metadata"] | CODE |
| LOW | GenAILab/qai_hub_lm/backends/torch/gemma4.py | 285 | def _set_qdq(owner, kind: str, idx: int, scale, num_bits: int) -> None: | CODE |
| LOW | GenAILab/qai_hub_lm/models/components.py | 32 | __all__ = [ | CODE |
| LOW | GenAILab/qai_hub_lm/schema/__init__.py | 64 | __all__ = [ | CODE |
| LOW | scripts/all/build_and_test.py | 71 | logger = logging.getLogger(__name__) | CODE |
| LOW | AIMETRegression/features/onnx/_common.py | 45 | __all__ = [ | CODE |
| LOW | AIMETRegression/features/torch/_common.py | 48 | __all__ = [ | CODE |
| LOW | AIMETRegression/workflow/__init__.py | 16 | __all__ = [ | CODE |
| LOW | AIMETRegression/workflow/artifacts.py | 533 | def _set_github_output(name: str, value: str) -> None: | CODE |
| LOW | AIMETRegression/evaluation/calibration.py | 29 | __all__ = [ | CODE |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/ir_utils.py | 33 | def set_static_tensor(value: onnx_ir.Value, array: np.ndarray) -> None: | CODE |
| LOW⚡ | …rc/python/aimet_onnx/experimental/lora/prepare_lora.py | 51 | logger = logging.getLogger(__name__) | CODE |
| LOW | …rc/python/aimet_onnx/experimental/lora/peft_to_onnx.py | 19 | logger = logging.getLogger(__name__) | CODE |
| LOW | …nx/src/python/aimet_onnx/experimental/lora/__init__.py | 94 | __all__ = [ | CODE |
| LOW⚡ | …et_onnx/experimental/lora/lora_adapter_quantization.py | 22 | logger = logging.getLogger(__name__) | CODE |
| LOW⚡ | …/python/aimet_onnx/experimental/lora/lora_configure.py | 24 | logger = logging.getLogger(__name__) | CODE |
| LOW | …thon/aimet_onnx/experimental/lora/lora_graph_passes.py | 21 | logger = logging.getLogger(__name__) | CODE |
| LOW | …x/src/python/aimet_onnx/llm_topology/norm_detection.py | 263 | __all__ = [ | CODE |
| LOW | …x/src/python/aimet_onnx/llm_topology/topology_types.py | 230 | __all__ = [ | CODE |
| LOW | …ns/onnx/src/python/aimet_onnx/llm_topology/__init__.py | 27 | __all__ = [ | CODE |
| LOW | …ns/onnx/src/python/aimet_onnx/llm_topology/topology.py | 993 | __all__ = [ | CODE |
| LOW | …src/python/aimet_onnx/llm_topology/block_boundaries.py | 292 | __all__ = [ | CODE |
| LOW | …onnx/src/python/aimet_onnx/llm_topology/layer_roles.py | 124 | __all__ = [ | CODE |
| LOW | …/onnx/src/python/aimet_onnx/llm_topology/ir_adapter.py | 366 | __all__ = [ | CODE |
| LOW | …onnx/src/python/aimet_onnx/llm_topology/hf_patterns.py | 682 | __all__ = [ | CODE |
| LOW | …nnx/src/python/aimet_onnx/llm_configurator/__init__.py | 8 | __all__ = ["configure_llm"] | CODE |
| LOW | …sions/onnx/src/python/aimet_onnx/spinquant/__init__.py | 7 | __all__ = [ | CODE |
| LOW | …nnx/src/python/aimet_onnx/spinquant/passes/__init__.py | 27 | __all__ = [ | CODE |
| LOW | …src/python/aimet_onnx/spinquant/transforms/__init__.py | 28 | __all__ = [ | CODE |
| LOW | …met_onnx/spinquant/model_analysis/attention_anchors.py | 407 | __all__ = ["BlockR3Anchors", "find_r3_anchors"] | CODE |
| LOW | …python/aimet_onnx/spinquant/model_analysis/__init__.py | 26 | __all__ = [ | CODE |
| LOW | …nsions/onnx/src/python/aimet_onnx/analysis/__init__.py | 22 | __all__ = [ | CODE |
| LOW | …nsions/onnx/src/python/aimet_onnx/adascale/__init__.py | 13 | __all__ = [ | CODE |
| LOW | …ningExtensions/torch/src/python/aimet_torch/seq_mse.py | 35 | __all__ = [ | CODE |
| LOW | …nsions/torch/src/python/aimet_torch/batch_norm_fold.py | 19 | __all__ = [ | CODE |
| LOW | …rch/src/python/aimet_torch/adaround/adaround_weight.py | 25 | __all__ = [ | CODE |
| LOW | …sions/torch/src/python/aimet_torch/experimental/pgs.py | 7 | __all__ = [ | CODE |
| LOW | …h/src/python/aimet_torch/experimental/export/export.py | 24 | __all__ = ["export"] | CODE |
| LOW | …ons/torch/src/python/aimet_torch/spinquant/__init__.py | 8 | __all__ = ["apply_spinquant"] | CODE |
| LOW | …ningExtensions/torch/src/python/aimet_torch/nn/lora.py | 10 | __all__ = [ | CODE |
| LOW | …ions/torch/src/python/aimet_torch/quantsim/quantsim.py | 77 | __all__ = [ | CODE |
| LOW | …ns/torch/src/python/aimet_torch/quantization/tensor.py | 20 | __all__ = [ | CODE |
| LOW | …src/python/aimet_torch/quantization/affine/encoding.py | 30 | __all__ = ["AffineEncoding", "VectorEncoding", "GroupedBlockEncoding"] | CODE |
| LOW | …rc/python/aimet_torch/quantization/affine/quantizer.py | 61 | __all__ = [ | CODE |
| LOW | …thon/aimet_torch/quantization/affine/backends/utils.py | 117 | __all__ = ["set_backend", "get_backend", "add_backend"] | CODE |
| LOW | …/src/python/aimet_torch/quantization/float/encoding.py | 21 | __all__ = ["FloatEncoding"] | CODE |
| LOW | …src/python/aimet_torch/quantization/float/quantizer.py | 41 | __all__ = ["QuantizeDequantize", "FloatQuantizeDequantize"] | CODE |
| LOW | …h/src/python/aimet_torch/quantization/base/encoding.py | 18 | __all__ = ["EncodingBase"] | CODE |
| LOW | …/src/python/aimet_torch/quantization/base/quantizer.py | 27 | __all__ = ["QuantizerBase"] | CODE |
| LOW | …ions/torch/src/python/aimet_torch/adascale/__init__.py | 9 | __all__ = ["apply_adascale"] | CODE |
| LOW | …rch/src/python/aimet_torch/_base/nn/modules/_spconv.py | 10 | __all__ = [ | CODE |
| LOW | …s/common/src/python/_aimet_common/hadamard/__init__.py | 17 | __all__ = [ | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/plugins/local/aimet.py | 199 | # For PyPi releases, just return the version without appending the variant string | COMMENT |
| MEDIUM | …Lab/bench/prompts/text_prompts_autograder_harness.yaml | 1 | # Autograder harness configuration for scoring LLM prompt responses. | COMMENT |
| MEDIUM | …nch/prompts/multimodal_prompts_autograder_harness.yaml | 1 | # Autograder harness configuration for scoring VLM prompt responses. | COMMENT |
| LOW⚡ | GenAILab/qai_hub_lm/backends/onnx/export_utils.py | 51 | # user has passed in a local path, verify that .onnx file and .config files exist and just return the path | COMMENT |
| MEDIUM | …b/qai_hub_lm/transforms/exportable_linear_attention.py | 616 | # specialization. That does not weaken the single-graph paradigm: both | COMMENT |
| LOW | Docs/snippets/onnx/apply_litemp.py | 114 | # If CUDA acceleration is not available, simply use providers = ["CPUExecutionProvider"] | COMMENT |
| MEDIUM | Examples/torch/compression/channel_pruning.py | 36 | # This script utilize AIMET to perform channel pruning compression (0.5% ratio) on a resnet18 | COMMENT |
| MEDIUM | Examples/torch/compression/weight_svd.py | 35 | # This script utilize AIMET do perform weight svd compression (50% ratio) on a resnet18 | COMMENT |
| MEDIUM | Examples/torch/compression/spatial_svd_cp.py | 39 | # This script utilize AIMET do perform spatial svd and channel pruning compression on a resnet18 | COMMENT |
| MEDIUM | Examples/torch/compression/spatial_svd.py | 35 | # This script utilize AIMET do perform spatial svd compression (50% compression ratio) on a | COMMENT |
| MEDIUM | …rc/python/aimet_onnx/experimental/lora/peft_to_onnx.py | 107 | # Build suffix → PEFT key index for robust matching. | COMMENT |
| MEDIUM | …sions/torch/test/python/experimental/test_spinquant.py | 164 | ) # hidden_size 192 will utilize hadamard matrix size 12 with factor 16 | CODE |
| LOW | …nsions/torch/src/python/aimet_torch/data_subsampler.py | 158 | # just return the input and output data as it is. No sub sampling needed | COMMENT |
| MEDIUM | …ainingExtensions/torch/src/python/aimet_torch/utils.py | 1936 | # Stage 1: robust median seed. | COMMENT |
| MEDIUM | …ainingExtensions/torch/src/python/aimet_torch/utils.py | 1903 | """Recover the per-block scale in two stages: a robust median seed followed by | STRING |
| MEDIUM | …ions/torch/src/python/aimet_torch/blockwise_sampler.py | 317 | """Exception raised in order to stop forward execution through the model. Holds module input data.""" | STRING |
| LOW | …gExtensions/torch/src/python/aimet_torch/onnx_utils.py | 208 | # 2) If the next token don't have "." then simply add current token to the updated_tokens list. | COMMENT |
| LOW⚡ | …s/common/src/python/_aimet_common/comp_ratio_select.py | 210 | # If the user already passed in a previously saved eval scores dict, we just use that | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | GenAILab/bench/metrics.py | 2121 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/internvl.py | 173 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/gemma4.py | 370 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/gemma4.py | 402 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/generator.py | 1015 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/generator.py | 1319 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/generator.py | 1451 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/gemma3.py | 108 | CODE | |
| LOW | GenAILab/qai_hub_lm/models/gemma3.py | 135 | CODE | |
| LOW | scripts/all/build_and_test.py | 214 | CODE | |
| LOW | TrainingExtensions/onnx/test/python/test_lora.py | 1989 | CODE | |
| LOW | …tensions/onnx/test/python/models/transformer_blocks.py | 496 | CODE | |
| LOW | …tensions/onnx/test/python/models/transformer_blocks.py | 552 | CODE | |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 207 | CODE | |
| LOW | …iningExtensions/onnx/src/python/aimet_onnx/quantsim.py | 426 | CODE | |
| LOW | …/src/python/aimet_onnx/graph_passes/fusions/rmsnorm.py | 139 | CODE | |
| LOW | …/src/python/aimet_onnx/graph_passes/fusions/rmsnorm.py | 255 | CODE | |
| LOW | …rc/python/aimet_onnx/graph_passes/fusions/layernorm.py | 166 | CODE | |
| LOW | …c/python/aimet_onnx/graph_passes/fusions/matmul_add.py | 126 | CODE | |
| LOW | …sions/onnx/src/python/aimet_onnx/adascale/quantizer.py | 236 | CODE | |
| LOW | …Extensions/torch/test/python/test_blockwise_sampler.py | 33 | CODE | |
| LOW | …sions/torch/test/python/experimental/test_omniquant.py | 173 | CODE | |
| LOW | …ningExtensions/torch/test/python/models/test_models.py | 1196 | CODE | |
| LOW | …ningExtensions/torch/test/python/models/test_models.py | 1220 | CODE | |
| LOW | …Extensions/torch/test/python/v2/models_/test_models.py | 1077 | CODE | |
| LOW | …Extensions/torch/test/python/v2/models_/test_models.py | 1101 | CODE | |
| LOW | TrainingExtensions/torch/src/python/aimet_torch/peft.py | 92 | CODE | |
| LOW | …tensions/torch/src/python/aimet_torch/nn/true_quant.py | 1354 | CODE | |
| LOW | …tensions/torch/src/python/aimet_torch/nn/true_quant.py | 1397 | CODE | |
| LOW | …c/python/aimet_torch/quantization/encoding_analyzer.py | 921 | CODE | |
| LOW | …src/python/aimet_torch/quantization/float/quantizer.py | 105 | CODE | |
| LOW | …orch/src/python/aimet_torch/_base/nn/modules/custom.py | 329 | CODE | |
| LOW | …orch/src/python/aimet_torch/_base/nn/modules/custom.py | 350 | CODE | |
| LOW | …orch/src/python/aimet_torch/_base/nn/modules/custom.py | 666 | CODE | |
| LOW | …orch/src/python/aimet_torch/_base/nn/modules/custom.py | 765 | CODE | |
| LOW | …rch/src/python/aimet_torch/_base/nn/modules/_spconv.py | 103 | CODE | |
| LOW | …rch/src/python/aimet_torch/_base/nn/modules/_spconv.py | 194 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …gacy/onnx_code_examples/quant_analyzer_code_example.py | 77 | dummy_data = np.random.randn(*input_shape).astype(np.float32) | CODE |
| LOW | …gacy/onnx_code_examples/quant_analyzer_code_example.py | 78 | dummy_input = {'input': dummy_data} | CODE |
| LOW | TrainingExtensions/onnx/test/python/test_quantsim.py | 8213 | name="placeholder", | CODE |
| LOW | …sions/torch/test/python/v2/experimental/test_export.py | 174 | assert inp_node.op == "placeholder" | CODE |
| LOW | …on/aimet_torch/experimental/export/exported_program.py | 82 | if node.op == "placeholder" | CODE |
| LOW | …on/aimet_torch/experimental/export/exported_program.py | 335 | if node.op != "placeholder": | CODE |
| LOW | …h/src/python/aimet_torch/experimental/export/export.py | 362 | and q_dq_node.all_input_nodes[1].op != "placeholder" | CODE |
| LOW | …h/src/python/aimet_torch/experimental/export/export.py | 376 | and q_dq_node.all_input_nodes[2].op != "placeholder" | CODE |
| LOW | …h/src/python/aimet_torch/experimental/export/_utils.py | 151 | for input_node in ep.graph.find_nodes(op="placeholder", sort=False) | CODE |
| LOW | …h/src/python/aimet_torch/experimental/export/_utils.py | 176 | op="placeholder", | CODE |
| LOW | …h/src/python/aimet_torch/experimental/export/_utils.py | 216 | if node.op == "placeholder": | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | packaging/convert_wheel_format.sh | 9 | # Usage: | COMMENT |
| LOW | scripts/all/download_genai_checkpoint.sh | 7 | # Usage: | COMMENT |
| LOW | scripts/all/resolve_genai_config.sh | 10 | # Usage: | COMMENT |
| LOW | scripts/environment/setup_genai.sh | 8 | # Usage: | COMMENT |
| LOW | scripts/environment/build_aimet.sh | 7 | # Usage: | COMMENT |
| LOW | .github/actions/genai-test/action.yml | 13 | # Usage: | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …ns/torch/src/python/aimet_torch/meta/connectedgraph.py | 1589 | # In this version of the Winnower, the Product class represents the following entities in a model. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | GenAILab/bench/datasets.py | 684 | # Build user prompt with optional few-shot examples | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | TrainingExtensions/onnx/test/python/test_quantsim.py | 6426 | # TODO: Add tests with GroupNormalization | COMMENT |
| LOW | …hon/aimet_torch/experimental/quantsim/_export_utils.py | 20 | # TODO: implement direct 1.0.0 export for v1 quantsim so this function can be removed. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …ions/onnx/test/python/test_llm_topology_integration.py | 500 | # architecture, make sure to extend the role module names in this dict. | COMMENT |
| LOW | …ions/torch/src/python/aimet_torch/quantsim/quantsim.py | 755 | # In case of any exception, make sure to restore | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …Lab/qai_hub_lm/backends/onnx/adaptations/hub_models.py | 4 | """ONNX-specific model adaptations to use checkpoints generated by AI Hub Models.""" | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …ingExtensions/common/src/python/_aimet_common/utils.py | 396 | >>> def do_something(): | STRING |