LiteRT, successor to TensorFlow Lite. is Google's On-device framework for high-performance ML & GenAI deployment on edge platforms, via efficient conversion, runtime, and optimization
This report summarizes scanner evidence for google-ai-edge/LiteRT, a C++ project with 3,485 GitHub stars. SynthScan examined 615,769 lines of code across 3532 source files, recording 6053 pattern matches distributed across 22 syntactic categories. The legacy adjusted pattern signal is 13.1 (Moderate 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? 4974 finding(s).
Are imports declared and package/API names verified? 464 finding(s).
Are wrappers, guards, and repeated structures necessary? 1794 finding(s).
Which changes need focused human review? 6053 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 6053 distinct pattern matches across 22 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 | configure.py | 1 | # Copyright 2024 The AI Edge LiteRT Authors. | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 1 | # | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 541 | # ml-dtypes | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 561 | # -r ci/official/requirements_updater/nvidia-constraints.txt | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 601 | # via | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 641 | # via | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 681 | # via | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 721 | --hash=sha256:ecd0a012051abc20c1aa87328841efa8cade3ced65803046e38c2f03c0891fea | COMMENT |
| LOW | requirements_lock_3_14_freethreaded.txt | 741 | # -r ci/official/requirements_updater/nvidia-requirements.txt | COMMENT |
| LOW | requirements_lock_3_14.txt | 1 | # | COMMENT |
| LOW | requirements_lock_3_14.txt | 541 | # ml-dtypes | COMMENT |
| LOW | requirements_lock_3_14.txt | 561 | # -r ci/official/requirements_updater/nvidia-constraints.txt | COMMENT |
| LOW | requirements_lock_3_14.txt | 601 | # via | COMMENT |
| LOW | requirements_lock_3_14.txt | 641 | # via | COMMENT |
| LOW | requirements_lock_3_14.txt | 681 | # via | COMMENT |
| LOW | requirements_lock_3_14.txt | 721 | --hash=sha256:8ce1b8213f61f2bfac132e6df890af6450b77cbd140c6ce4e98cb0c2d8e678c9 | COMMENT |
| LOW | requirements_lock_3_14.txt | 741 | # nvidia-cufft | COMMENT |
| LOW | requirements_lock_3_13.txt | 1 | # | COMMENT |
| LOW | requirements_lock_3_13.txt | 441 | --hash=sha256:e6ecde441aaf0bb74ed538cfb3b18aa374f452aebf0162088bcb10942f7bbc33 \ | COMMENT |
| LOW | requirements_lock_3_13.txt | 601 | # -r ci/official/requirements_updater/nvidia-constraints.txt | COMMENT |
| LOW | requirements_lock_3_13.txt | 621 | --hash=sha256:ca786ffa5a647c75d4d1f5cc72a6c4f537947e2ba8823d7c8aaf768e7a7b9f77 \ | COMMENT |
| LOW | requirements_lock_3_13.txt | 641 | --hash=sha256:f9b37bc5c8cf7509665cb6ada5aaa0ce65618f2332b7d3e78e9790511f111212 | COMMENT |
| LOW | requirements_lock_3_12.txt | 1 | # | COMMENT |
| LOW | requirements_lock_3_12.txt | 441 | --hash=sha256:e6ecde441aaf0bb74ed538cfb3b18aa374f452aebf0162088bcb10942f7bbc33 \ | COMMENT |
| LOW | requirements_lock_3_12.txt | 601 | # -r ci/official/requirements_updater/nvidia-constraints.txt | COMMENT |
| LOW | requirements_lock_3_12.txt | 621 | --hash=sha256:ca786ffa5a647c75d4d1f5cc72a6c4f537947e2ba8823d7c8aaf768e7a7b9f77 \ | COMMENT |
| LOW | requirements_lock_3_12.txt | 641 | --hash=sha256:f9b37bc5c8cf7509665cb6ada5aaa0ce65618f2332b7d3e78e9790511f111212 | COMMENT |
| LOW | requirements_lock_3_12.txt | 1081 | --hash=sha256:29b23c360f22f414dc7336bb39178cc7bcbf6021ed2733cde173f09dba19abb3 | COMMENT |
| LOW | requirements_lock_3_10.txt | 1 | # | COMMENT |
| LOW | requirements_lock_3_10.txt | 441 | --hash=sha256:e6ecde441aaf0bb74ed538cfb3b18aa374f452aebf0162088bcb10942f7bbc33 \ | COMMENT |
| LOW | requirements_lock_3_10.txt | 601 | # -r ci/official/requirements_updater/nvidia-constraints.txt | COMMENT |
| LOW | requirements_lock_3_10.txt | 621 | --hash=sha256:ca786ffa5a647c75d4d1f5cc72a6c4f537947e2ba8823d7c8aaf768e7a7b9f77 \ | COMMENT |
| LOW | requirements_lock_3_10.txt | 641 | --hash=sha256:f9b37bc5c8cf7509665cb6ada5aaa0ce65618f2332b7d3e78e9790511f111212 | COMMENT |
| LOW | requirements_lock_3_11.txt | 1 | # | COMMENT |
| LOW | requirements_lock_3_11.txt | 441 | --hash=sha256:e6ecde441aaf0bb74ed538cfb3b18aa374f452aebf0162088bcb10942f7bbc33 \ | COMMENT |
| LOW | requirements_lock_3_11.txt | 601 | # -r ci/official/requirements_updater/nvidia-constraints.txt | COMMENT |
| LOW | requirements_lock_3_11.txt | 621 | --hash=sha256:ca786ffa5a647c75d4d1f5cc72a6c4f537947e2ba8823d7c8aaf768e7a7b9f77 \ | COMMENT |
| LOW | requirements_lock_3_11.txt | 641 | --hash=sha256:f9b37bc5c8cf7509665cb6ada5aaa0ce65618f2332b7d3e78e9790511f111212 | COMMENT |
| LOW | requirements_lock_3_11.txt | 1081 | --hash=sha256:29b23c360f22f414dc7336bb39178cc7bcbf6021ed2733cde173f09dba19abb3 | COMMENT |
| LOW | Package.swift | 1 | // swift-tools-version: 5.9 | COMMENT |
| LOW | …t_ats_op_generator/references/primitive_op_patterns.md | 221 | #include <type_traits> | COMMENT |
| LOW | tflite/generate-pc.sh | 1 | #!/usr/bin/env bash | COMMENT |
| LOW | tflite/version.h | 21 | // semantic versioning https://semver.org/ | COMMENT |
| LOW | tflite/version.h | 41 | TF_PATCH_VERSION) TF_VERSION_SUFFIX) | COMMENT |
| LOW | tflite/portable_type_to_tflitetype.h | 21 | COMMENT | |
| LOW | tflite/CMakeLists.txt | 1 | # | COMMENT |
| LOW | tflite/CMakeLists.txt | 101 | set(_TFLITE_ENABLE_NNAPI OFF) | COMMENT |
| LOW | tflite/minimal_logging.h | 81 | #define TFLITE_LOG TFLITE_LOG_PROD | COMMENT |
| LOW | tflite/string_util.h | 21 | // [(N+1)*4, (N+1)*4+3] 4 bytes: length of the whole char buffer. | COMMENT |
| LOW | tflite/arena_planner.h | 21 | #include <unordered_map> | COMMENT |
| LOW | tflite/arena_planner.h | 41 | // share some of the buffer if a tensor B is to be allocated after another | COMMENT |
| LOW | tflite/arena_planner.h | 81 | int output_id, bool tensor_changed); | COMMENT |
| LOW | tflite/memory_planner.h | 41 | // ExecuteAllocations() is called. | COMMENT |
| LOW | tflite/namespace.h | 21 | // To avoid potential violation of the C++ "one definition rule" (ODR) for | COMMENT |
| LOW | tflite/interpreter_options.h | 41 | /// inference speed. | COMMENT |
| LOW | tflite/interpreter_options.h | 121 | // Controls to update Tensor names with Signature input & output names. | COMMENT |
| LOW | tflite/simple_memory_arena.h | 121 | COMMENT | |
| LOW | tflite/simple_memory_arena.h | 161 | COMMENT | |
| LOW | tflite/external_cpu_backend_context.h | 41 | }; | COMMENT |
| LOW | tflite/external_cpu_backend_context.h | 61 | // | COMMENT |
| 3271 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | configure.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/generate-pc.sh | 14 | # ============================================================================== | COMMENT |
| MEDIUM | …eleration/compatibility/convert_binary_to_cc_source.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | …end/python/kernel_tests/audio_microfrontend_op_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | …tal/microfrontend/python/ops/audio_microfrontend_op.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/convert_image_to_csv.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/build_aar.sh | 15 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/test_utils.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/convert_image_to_csv_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/randomize_weights.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/zip_files.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/strip_strings.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/visualize.py | 15 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/build_aar_with_docker.sh | 15 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/reverse_xxd_dump_from_cc.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/visualize_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | …ize/sparsity/format_converter_wrapper_pybind11_test.py | 14 | # ============================================================================= | COMMENT |
| MEDIUM | …ols/optimize/python/modify_model_interface_lib_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | …te/tools/optimize/python/modify_model_interface_lib.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | …ls/optimize/python/modify_model_interface_constants.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/optimize/python/modify_model_interface.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/optimize/debugging/python/debugger_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/optimize/debugging/python/debugger.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/cmake/download_toolchains.sh | 15 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/pip_package/setup_with_binary.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/tools/pip_package/update_sources.sh | 15 | # ============================================================================== | COMMENT |
| MEDIUM | …lite/tools/pip_package/build_pip_package_with_cmake.sh | 15 | # ============================================================================== | COMMENT |
| MEDIUM | …lite/tools/pip_package/build_pip_package_with_bazel.sh | 15 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/converter/python/wrap_converter.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/convert_saved_model.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/convert_file_to_c_source_test.sh | 15 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/schema_util.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/wrap_toco.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/convert_saved_model_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/util.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/op_hint.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/convert.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/lite_v2_test_util.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM⚡ | tflite/python/analyzer.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/convert_phase.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/lite_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/test_util.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/util_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/interpreter.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/interpreter_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/lite_flex_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/analyzer_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/lite.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/test_util_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/tflite_keras_util.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/convert_file_to_c_source.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/convert_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/lite_constants.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/metrics/metrics_nonportable_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM | tflite/python/metrics/metrics_portable_test.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM⚡ | tflite/python/metrics/metrics_interface.py | 14 | # ============================================================================== | COMMENT |
| MEDIUM⚡ | tflite/python/metrics/metrics_portable.py | 14 | # ============================================================================== | COMMENT |
| 460 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | configure.py | 114 | def write_action_env_to_bazelrc(var_name, var): | CODE |
| LOW | configure.py | 118 | def write_repo_env_to_bazelrc(config_name, var_name, var): | CODE |
| LOW | configure.py | 255 | def reset_tf_configure_bazelrc(): | CODE |
| LOW | configure.py | 535 | def get_from_env_or_user_or_default( | CODE |
| LOW | configure.py | 561 | def prompt_loop_or_load_from_env( | CODE |
| LOW | configure.py | 631 | def set_clang_cuda_compiler_path(environ_cp): | CODE |
| LOW | configure.py | 822 | def set_gcc_host_compiler_path(environ_cp): | CODE |
| LOW | configure.py | 908 | def set_clang_compiler_path_win(environ_cp): | CODE |
| LOW | configure.py | 984 | def disable_clang_offsetof_extension(clang_version): | CODE |
| LOW | configure.py | 989 | def set_hermetic_cuda_version(environ_cp): | CODE |
| LOW | configure.py | 1005 | def set_hermetic_cudnn_version(environ_cp): | CODE |
| LOW | configure.py | 1021 | def set_hermetic_cuda_compute_capabilities(environ_cp): | CODE |
| LOW | configure.py | 1128 | def system_specific_test_config(environ_cp): | CODE |
| LOW | …eleration/compatibility/convert_binary_to_cc_source.py | 30 | def _convert_bytes_to_cc_source(data, | CODE |
| LOW | tflite/tools/test_utils.py | 26 | def build_mock_flatbuffer_model(): | CODE |
| LOW | tflite/tools/test_utils.py | 273 | def build_operator_with_options() -> schema_fb.Operator: | CODE |
| LOW | tflite/tools/test_utils.py | 289 | def load_model_from_flatbuffer(flatbuffer_model): | CODE |
| LOW | tflite/tools/flatbuffer_utils.py | 39 | def convert_bytearray_to_object(model_bytearray): | CODE |
| LOW | tflite/tools/flatbuffer_utils.py | 65 | def read_model_from_bytearray(model_bytearray): | CODE |
| LOW | tflite/tools/flatbuffer_utils.py | 97 | def read_model_with_mutable_tensors(input_tflite_file): | CODE |
| LOW | tflite/tools/flatbuffer_utils.py | 119 | def convert_object_to_bytearray(model_object, extra_buffer=b''): | CODE |
| LOW | tflite/tools/flatbuffer_utils.py | 358 | def byte_swap_tflite_model_obj(model, from_endiness, to_endiness): | CODE |
| LOW⚡ | tflite/tools/flatbuffer_utils_test.py | 278 | def test_get_options_not_schema_type_raises(self): | CODE |
| LOW⚡ | tflite/tools/flatbuffer_utils_test.py | 282 | def test_get_options_not_object_type_raises(self): | CODE |
| LOW⚡ | tflite/tools/flatbuffer_utils_test.py | 286 | def test_get_options_op_type_does_not_match(self): | CODE |
| LOW | …ols/optimize/python/modify_model_interface_lib_test.py | 27 | def build_tflite_model_with_full_integer_quantization( | CODE |
| LOW | …ols/optimize/python/modify_model_interface_lib_test.py | 41 | def representative_dataset_gen(): | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger_test.py | 258 | def test_layer_direct_compare_metrics(self, quantized_io): | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger_test.py | 348 | def test_denylisted_ops_from_option_setter(self, quantized_io): | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger_test.py | 373 | def test_denylisted_ops_from_option_constructor(self, quantized_io): | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger_test.py | 392 | def test_denylisted_nodes_from_option_setter(self, quantized_io): | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger_test.py | 413 | def test_denylisted_nodes_from_option_constructor(self, quantized_io): | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 247 | def get_nondebug_quantized_model(self) -> bytes: | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 261 | def get_debug_quantized_model(self) -> bytes: | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 306 | def _set_converter_options_for_float( | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 313 | def _set_converter_options_for_calibration( | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 332 | def _collect_layer_statistics(self) -> Dict[str, Dict[str, float]]: | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 386 | def _collect_model_statistics(self) -> Dict[str, float]: | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 490 | def _get_numeric_verify_tensor_details(self) -> List[str]: | CODE |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 506 | def _get_operand_name_and_index(self, | CODE |
| LOW | tflite/converter/python/wrap_converter.py | 40 | def wrapped_convert_mlir_bytecode( | CODE |
| LOW | tflite/converter/python/wrap_converter.py | 53 | def wrapped_experimental_mlir_quantize( | CODE |
| LOW⚡ | tflite/converter/python/wrap_converter.py | 86 | def wrapped_experimental_mlir_sparsify(input_data_str): | CODE |
| LOW⚡ | tflite/converter/python/wrap_converter.py | 91 | def wrapped_register_custom_opdefs(custom_opdefs_list): | CODE |
| LOW⚡ | tflite/converter/python/wrap_converter.py | 96 | def wrapped_retrieve_collected_errors(): | CODE |
| LOW⚡ | tflite/converter/python/wrap_converter.py | 101 | def wrapped_flat_buffer_file_to_mlir( | CODE |
| LOW | tflite/converter/python/wrap_converter.py | 115 | def wrapped_mlir_to_flat_buffer( | CODE |
| LOW | tflite/python/schema_util.py | 20 | def get_builtin_code_from_operator_code(opcode): | CODE |
| LOW⚡ | tflite/python/util.py | 568 | def get_dequantize_opcode_idx(model): | CODE |
| LOW⚡ | tflite/python/util.py | 578 | def _update_signature_def_tensors(tensor_maps, map_old_to_new_tensors): | CODE |
| LOW⚡ | tflite/python/util.py | 586 | def _remove_tensors_from_model(model, remove_tensors_idxs): | CODE |
| LOW | tflite/python/util.py | 84 | def _convert_tflite_enum_type_to_tf_type(tflite_enum_type): | CODE |
| LOW | tflite/python/util.py | 130 | def get_tensors_from_tensor_names(graph, tensor_names): | CODE |
| LOW | tflite/python/util.py | 262 | def _convert_op_hints_if_present(sess, graph_def, output_tensors, | CODE |
| LOW | tflite/python/util.py | 421 | def convert_bytes_to_c_source(data, | CODE |
| LOW | tflite/python/util.py | 549 | def _convert_model_from_object_to_bytearray(model_object): | CODE |
| LOW | tflite/python/util.py | 649 | def _modify_model_input_type_per_subgraph(model, subgraph_index, | CODE |
| LOW | tflite/python/util.py | 741 | def _modify_model_output_type(model, inference_output_type=dtypes.float32): | CODE |
| LOW | tflite/python/util.py | 756 | def _modify_model_output_type_per_subgraph(model, subgraph_index, | CODE |
| LOW | tflite/python/util.py | 864 | def _remove_redundant_quantize_ops(model): | CODE |
| 1086 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | configure.py | 284 | Get boolean input from user. If var_name is not set in env, ask user to enable query_item or not. If the response i | STRING |
| HIGH | configure.py | 572 | Loop over user prompts for an ENV param until receiving a valid response. For the env param var_name, read from the e | STRING |
| HIGH | …tal/microfrontend/python/ops/audio_microfrontend_op.py | 51 | Audio Microfrontend Op. This Op converts a sequence of audio data into one or more feature vectors containing filte | STRING |
| HIGH | tflite/tools/flatbuffer_utils.py | 46 | Reads a tflite model as a python object. Args: input_tflite_file: Full path name to the input tflite file Rais | STRING |
| HIGH | tflite/tools/flatbuffer_utils.py | 98 | Reads a tflite model as a python object with mutable tensors. Similar to read_model() with the addition that the retu | STRING |
| HIGH | tflite/tools/flatbuffer_utils.py | 272 | Converts xxd output C++ source file to bytes (immutable). Args: input_cc_file: Full path name to th C++ source fi | STRING |
| HIGH | tflite/tools/flatbuffer_utils.py | 311 | Converts xxd output C++ source file to object. Args: input_cc_file: Full path name to th C++ source file dumped b | STRING |
| HIGH | tflite/tools/flatbuffer_utils.py | 471 | Get the options of an operator as the specified type. Requested type must be an object-api type (ends in 'T'). Arg | STRING |
| HIGH | tflite/tools/visualize.py | 461 | Returns html description with the given tflite model. Args: tflite_input: TFLite flatbuffer model path or model o | STRING |
| HIGH | …te/tools/optimize/python/modify_model_interface_lib.py | 23 | Converts a tflite type to it's integer representation. Args: dtype: tf.DType representing the inference type. | STRING |
| HIGH | tflite/tools/optimize/debugging/python/debugger.py | 138 | Runs the TFLite debugging model with given debug options. Args: quant_debug_model_path: Path to the quantized | STRING |
| HIGH | tflite/python/convert_saved_model.py | 29 | Validate saved_model and extract MetaGraphDef. Args: saved_model_dir: saved_model path to convert. tag_set: S | STRING |
| HIGH | tflite/python/convert_saved_model.py | 46 | Get the signature def from meta_graph with given signature_key. Args: meta_graph: meta_graph_def. signature_k | STRING |
| HIGH | tflite/python/convert_saved_model.py | 92 | Gets the tensors associated with the tensor names. Either signature_def_tensor_names or user_tensor_names should be p | STRING |
| HIGH | tflite/python/convert_saved_model.py | 136 | Converts a SavedModel to a frozen graph. Args: saved_model_dir: SavedModel directory to convert. input_arrays | STRING |
| HIGH | tflite/python/util.py | 85 | Converts tflite enum type (eg: 0) to tf type (eg: tf.float32). Args: tflite_enum_type: tflite enum type (eg: 0, t | STRING |
| HIGH | tflite/python/util.py | 131 | Gets the Tensors associated with the `tensor_names` in the provided graph. Args: graph: TensorFlow Graph. ten | STRING |
| HIGH | tflite/python/util.py | 944 | Modify the input/output type of a tflite model. Args: model: A tflite model. inference_input_type: tf.DType r | STRING |
| HIGH | tflite/python/op_hint.py | 200 | Return the next unused argument index in order or use an override. Args: index_override: An index to use | STRING |
| HIGH | tflite/python/op_hint.py | 226 | Return a wrapped tensor of an input tensor as an argument. Args: arg: A TensorFlow tensor that should be | STRING |
| HIGH | tflite/python/op_hint.py | 480 | This adds the node(s) to out_graphdef and returns the input node name. Args: out_graphdef: A graphdef that is | STRING |
| HIGH | tflite/python/op_hint.py | 496 | Add node(s) to graph representing output operands and returns type. Args: fused_op_name: name of the fused op | STRING |
| HIGH | tflite/python/op_hint.py | 1246 | Find all Ophints output nodes in the graph. This is used to get all the output nodes those are ophinted, it is import | STRING |
| HIGH | tflite/python/op_hint.py | 1302 | Converts a graphdef with LiteOp hints into stub operations. This is used to prepare for toco conversion of complex in | STRING |
| HIGH | tflite/python/convert.py | 67 | Convert tensor type from tf type to tflite type. Args: tf_type: TensorFlow type. usage: Text describing the r | STRING |
| HIGH | tflite/python/convert.py | 111 | Convert inference type from tf type to tflite type. Args: tf_type: TensorFlow type. usage: Text describing th | STRING |
| HIGH | tflite/python/convert.py | 361 | Converts `input_data_str` to a TFLite model. Args: model_flags: Proto describing model properties, see `model_fla | STRING |
| HIGH | tflite/python/convert.py | 417 | Converts `model_dir` to a TFLite model file directly. Args: conversion_flags: Proto describing conversion propert | STRING |
| HIGH | tflite/python/convert.py | 550 | Builds protocol buffer describing a conversion of a model. Typically this is to convert from TensorFlow GraphDef to T | STRING |
| HIGH | tflite/python/convert.py | 850 | Convert a frozen GraphDef that can't be loaded in TF. Conversion can be customized by providing arguments that are fo | STRING |
| HIGH | tflite/python/convert.py | 916 | Convert a frozen GraphDef model using the TF Lite converter. Conversion can be customized by providing arguments that | STRING |
| HIGH | tflite/python/convert.py | 1046 | Convert a TensorFlow GraphDef to TFLite. This function is deprecated. Please use `tf.lite.TFLiteConverter` API instea | STRING |
| HIGH | tflite/python/convert_phase.py | 166 | The decorator to identify converter component and subcomponent. Args: component: Converter component name. su | STRING |
| HIGH | tflite/python/interpreter.py | 139 | Returns loaded Delegate object. Example usage: ``` import tensorflow as tf try: delegate = tf.lite.experi | STRING |
| HIGH | tflite/python/interpreter.py | 637 | Gets tensor details. Args: tensor_index: Tensor index of tensor to query. subgraph_index: Index of the | STRING |
| HIGH | tflite/python/interpreter.py | 889 | Gets callable for inference of specific SignatureDef. Example usage, ``` interpreter = tf.lite.Interpreter( | STRING |
| HIGH | tflite/python/lite.py | 1482 | Converts a TensorFlow GraphDef based on instance variables. Args: graph_def: Frozen TensorFlow GraphDef. | STRING |
| HIGH | tflite/python/lite.py | 2262 | Creates a TFLiteConverter object from ConcreteFunctions. Args: funcs: List of TensorFlow ConcreteFunctions. T | STRING |
| HIGH | tflite/python/lite.py | 2304 | Creates a TFLiteConverter object from a SavedModel directory. Args: saved_model_dir: SavedModel directory to | STRING |
| HIGH | tflite/python/lite.py | 3274 | Creates a TFLiteConverter class from a file containing a frozen GraphDef. Args: graph_def_file: Full filepath | STRING |
| HIGH | tflite/python/tflite_keras_util.py | 193 | Trace the model call to create a tf.function for exporting a Keras model. Args: model: A Keras model. input_s | STRING |
| HIGH | tflite/testing/zip_test_utils.py | 260 | Given a dictionary of `examples`, write a text format representation. The file format is protocol-buffer-like, even t | STRING |
| HIGH | tflite/testing/zip_test_utils.py | 375 | Helper to make a zip file of a bunch of TensorFlow models. This does a cartesian product of the dictionary of test_pa | STRING |
| HIGH | tflite/schema/upgrade_schema.py | 103 | Read a tflite model assuming the given flatbuffer schema. If `input_file` is in bin, then we must use flatc to conv | STRING |
| HIGH | tflite/schema/upgrade_schema.py | 229 | Remap operator structs from old names to new names. Args: operator_type: String representing the builtin | STRING |
| HIGH | litert_quantizer/litert_quantizer/params_generator.py | 75 | Generate the quantization parameters for the model. Args: model_recipe_manager: The recipe manager for the mo | STRING |
| HIGH | litert_quantizer/litert_quantizer/params_generator.py | 336 | Check buffer sharing for the tensor against itself. Args: tensor: The tensor to check. Returns: Wh | STRING |
| HIGH | litert_quantizer/litert_quantizer/params_generator.py | 369 | Checks that self-compatible tensors have the same quantization params. Self-compatible means that all tensor's cons | STRING |
| HIGH | …rt_quantizer/litert_quantizer/algorithm_manager_api.py | 301 | Returns the list of supported ops for the given algorithm. Args: alg_key: Algorithm key. Returns: | STRING |
| HIGH | litert_quantizer/litert_quantizer/quantizer.py | 376 | Calibrates the float model (required by static range quantization). Args: calibration_data: Calibration data | STRING |
| HIGH | litert_quantizer/litert_quantizer/quantizer.py | 448 | Quantizes the float model. Also prints a report summarizing the quantization process after the process is done. | STRING |
| HIGH | litert_quantizer/litert_quantizer/calibrator.py | 220 | Creates a new instance of the appropriate calibrator subclass. Args: float_tflite: The path to the TFLite mod | STRING |
| HIGH | …tert_quantizer/transformation_instruction_generator.py | 705 | Split the instructions into subsets by tensor duplication. Splits the instructions into subsets based on which tens | STRING |
| HIGH | …er/litert_quantizer/transformations/insert_multiply.py | 26 | Inserts an elementwise MUL op on this tensor for activation scaling. This function works for float32 tensors. It inse | STRING |
| HIGH | …_quantizer/transformations/insert_hadamard_rotation.py | 63 | Inserts a custom aeq.hadamard_rotation op on this tensor. This function works for float32 tensors only. Args: | STRING |
| HIGH | …transformations/insert_decomposed_hadamard_rotation.py | 61 | Generates a Hadamard matrix of the given size. Args: size: The size of the Hadamard matrix. Must be a power of 2. | STRING |
| HIGH | …transformations/insert_decomposed_hadamard_rotation.py | 87 | Inserts a decomposed pattern of Hadamard rotation on this tensor. This function works for float32 tensors only. Inste | STRING |
| HIGH | …/litert_quantizer/algorithms/uniform_quantize/oscar.py | 466 | Returns the quantization parameters for a tensor. For weight tensors, replaces the max-abs bound with OSCAR's activ | STRING |
| HIGH | …tizer/algorithms/uniform_quantize/hadamard_rotation.py | 49 | Generates a Hadamard matrix of the given size. Hadamard matrices are square matrices with entries +1 or -1 and orthog | STRING |
| HIGH | …tizer/algorithms/uniform_quantize/hadamard_rotation.py | 98 | Quantizes the given float array using the diagonal Hadamard algorithm. Args: tensor_content: The float array to q | STRING |
| 44 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tflite/converter/python/wrap_converter.py | 18 | CODE | |
| LOW | tflite/python/wrap_toco.py | 18 | CODE | |
| LOW | tflite/python/lite.py | 34 | CODE | |
| LOW | tflite/python/lite.py | 37 | CODE | |
| LOW | tflite/python/lite.py | 38 | CODE | |
| LOW | tflite/python/lite.py | 49 | CODE | |
| LOW | tflite/python/lite.py | 52 | CODE | |
| LOW | tflite/python/lite.py | 54 | CODE | |
| LOW | tflite/python/lite.py | 59 | CODE | |
| LOW | tflite/python/lite.py | 60 | CODE | |
| LOW | tflite/python/lite.py | 61 | CODE | |
| LOW | tflite/python/lite.py | 63 | CODE | |
| LOW | tflite/python/lite.py | 65 | CODE | |
| LOW | tflite/python/lite.py | 87 | CODE | |
| LOW | tflite/python/lite.py | 88 | CODE | |
| LOW | tflite/python/metrics/wrapper/metrics_wrapper.py | 19 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 41 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 42 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 43 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 44 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 45 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 46 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 47 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 48 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 49 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 50 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 51 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 52 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 53 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 54 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 55 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 56 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 57 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 58 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 59 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 60 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 61 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 62 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 63 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 64 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 64 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 64 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 65 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 66 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 67 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 68 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 69 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 71 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 72 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 73 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 74 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 75 | CODE | |
| LOW | tflite/testing/generate_examples_lib.py | 76 | CODE | |
| 374 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tflite/tools/visualize.py | 32 | # This file is part of tensorflow package. | COMMENT |
| MEDIUM | tflite/tools/visualize.py | 35 | # This file is part of tflite_runtime package. | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils_test.py | 34 | # Define the initial model | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils_test.py | 79 | # Define the initial model | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils_test.py | 127 | # Define the initial model | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils_test.py | 167 | # Define the initial model | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils_test.py | 211 | # Define the initial model | COMMENT |
| MEDIUM | tflite/tools/flatbuffer_utils_test.py | 243 | # Define the initial model | COMMENT |
| MEDIUM | …ols/optimize/python/modify_model_interface_lib_test.py | 56 | # Define the temporary directory and files | COMMENT |
| MEDIUM | …ols/optimize/python/modify_model_interface_lib_test.py | 91 | # Define the temporary directory and files | COMMENT |
| MEDIUM | …ols/optimize/python/modify_model_interface_lib_test.py | 128 | # Define the temporary directory and files | COMMENT |
| MEDIUM | tflite/tools/optimize/debugging/python/debugger_test.py | 75 | # Create a TFLite model with new quantizer and numeric verify ops. | COMMENT |
| MEDIUM | tflite/python/util.py | 828 | # Create a quantize operator, if none exist | COMMENT |
| MEDIUM | tflite/python/op_hint.py | 1022 | # Create the function | COMMENT |
| MEDIUM | tflite/python/convert.py | 1172 | # Create a map of duplicate buffers (same size and same type). | COMMENT |
| MEDIUM⚡ | tflite/python/analyzer.py | 22 | # This file is part of tensorflow package. | COMMENT |
| MEDIUM⚡ | tflite/python/analyzer.py | 27 | # This file is part of tflite_runtime package. | COMMENT |
| MEDIUM | tflite/python/util_test.py | 229 | # Define a pseudo MNIST dataset (as downloading the dataset on-the-fly causes | COMMENT |
| MEDIUM | tflite/python/interpreter.py | 28 | # This file is part of tensorflow package. | COMMENT |
| MEDIUM | tflite/python/interpreter.py | 33 | # This file is part of tflite_runtime package. | COMMENT |
| MEDIUM | tflite/python/interpreter.py | 44 | # This file is part of the ai_edge_litert package. | COMMENT |
| MEDIUM | tflite/python/lite_flex_test.py | 78 | # Create a graph that has one L2Loss op. | COMMENT |
| MEDIUM | tflite/python/lite_flex_test.py | 212 | # Create a graph that has one add op. | COMMENT |
| MEDIUM | tflite/python/lite_flex_test.py | 254 | # Create a graph that has one double op. | COMMENT |
| MEDIUM⚡ | tflite/python/lite_v2_test.py | 1289 | # Create a TFLite model with new quantizer. | COMMENT |
| MEDIUM⚡ | tflite/python/lite_v2_test.py | 1293 | # Create a TFLite model with new quantizer and numeric verify ops. | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 1887 | # Create a v1 saved model with hash table initializers. | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 1976 | # Create a v1 saved model with mutable hash table. | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 3315 | # Create a simple Keras model. | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 3347 | # Create a simple Keras model. | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 3388 | # Create a simple Keras model. | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 5170 | # Create a graph that has one double op. | COMMENT |
| MEDIUM | tflite/python/metrics/metrics_nonportable_test.py | 436 | # Create a graph that has one add op. | COMMENT |
| MEDIUM⚡ | tflite/python/metrics/metrics_portable.py | 22 | # This file is part of tensorflow package. | COMMENT |
| MEDIUM⚡ | tflite/python/metrics/metrics_portable.py | 25 | # This file is part of tflite_runtime package. | COMMENT |
| MEDIUM⚡ | tflite/python/optimize/calibrator_test.py | 248 | # Define a simple model to run calibration with. | COMMENT |
| MEDIUM | tflite/testing/op_tests/fully_connected_4bit_hybrid.py | 52 | # Create a float filter with no quantization loss. | COMMENT |
| MEDIUM | litert_quantizer/litert_quantizer/litert_quantizer.py | 94 | # Create the output directory if it doesn't already exist. | COMMENT |
| MEDIUM | litert_quantizer/litert_quantizer/litert_quantizer.py | 139 | # Create a filename for the quantized TFLite model. | COMMENT |
| MEDIUM | litert_quantizer/litert_quantizer/litert_quantizer.py | 208 | # Create the output directory if it doesn't already exist. | COMMENT |
| MEDIUM | litert_quantizer/litert_quantizer/calibrator_test.py | 272 | # Create a legacy format file (just the dict) | COMMENT |
| MEDIUM | litert_quantizer/litert_quantizer/calibrator_test.py | 291 | # Create a file with custom metadata | COMMENT |
| MEDIUM⚡ | …tert_quantizer/transformations/transformation_utils.py | 153 | # Create a new `qtyping.BufferT` object. | COMMENT |
| MEDIUM | …er/algorithms/uniform_quantize/common_quantize_test.py | 167 | # Create a mock tfl_op with inputs and outputs | COMMENT |
| MEDIUM | …litert_quantizer/algorithms/utils/common_utils_test.py | 503 | # Create a mock op with multiple outputs. | COMMENT |
| MEDIUM | …ntizer/litert_quantizer/utils/constrained_ops_utils.py | 85 | # Create a dummy op info to trigger the materialization. | COMMENT |
| MEDIUM | litert_quantizer/litert_quantizer/utils/recipe_utils.py | 38 | # Create a mapping of recipe names to parsed JSON files. | COMMENT |
| MEDIUM | …ert_quantizer/litert_quantizer/utils/litertlm_utils.py | 235 | # Create a memory-mapped file in which we will pack the LiteRTLM file. | COMMENT |
| MEDIUM⚡ | …_quantizer/litert_quantizer/utils/recipe_utils_test.py | 72 | # Create a copy of the original recipe and mark it as such. | COMMENT |
| MEDIUM⚡ | …_quantizer/litert_quantizer/utils/recipe_utils_test.py | 75 | # Create an output directory outside of `cwd` and `recipes/` that will | COMMENT |
| MEDIUM⚡ | …_quantizer/litert_quantizer/utils/recipe_utils_test.py | 82 | # Create a litertlm recipe mapping. | COMMENT |
| MEDIUM | litert/support_lib/metadata/python/metadata.py | 159 | # Creating the deleting such a tempfile is handled by the class, | COMMENT |
| MEDIUM | litert/support_lib/metadata/python/metadata.py | 793 | # Create an individual method for getting the metadata json file, so that it can | COMMENT |
| MEDIUM | …ert/support_lib/metadata/python/tests/metadata_test.py | 66 | # Create a model with two inputs and one output, which matches the metadata | COMMENT |
| MEDIUM | …ert/support_lib/metadata/python/tests/metadata_test.py | 100 | # Create a model with two inputs and one output. | COMMENT |
| MEDIUM | …ert/support_lib/metadata/python/tests/metadata_test.py | 429 | # Create a model with two inputs and one output. | COMMENT |
| MEDIUM | …ert/support_lib/metadata/python/tests/metadata_test.py | 493 | # Create a dummy metadata without Subgraph. | COMMENT |
| MEDIUM | …ert/support_lib/metadata/python/tests/metadata_test.py | 509 | # Create a dummy metadata with no input tensor metadata, while the expected | COMMENT |
| MEDIUM | …ert/support_lib/metadata/python/tests/metadata_test.py | 530 | # Create a dummy metadata with no output tensor metadata, while the expected | COMMENT |
| MEDIUM | …ert/support_lib/metadata/python/tests/metadata_test.py | 551 | # Create a src model with metadata and two associated files. | COMMENT |
| 26 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL⚡ | tflite/python/lite.py | 3189 | session. converter = tf.compat.v1.lite.TFLiteConverter.from_session( sess, | STRING |
| CRITICAL⚡ | tflite/python/lite.py | 3193 | tf.compat.v1.lite.TFLiteConverter.from_frozen_graph( graph_def_file, | STRING |
| CRITICAL⚡ | tflite/python/lite.py | 3197 | tf.compat.v1.lite.TFLiteConverter.from_saved_model( saved_model_dir) | STRING |
| CRITICAL⚡ | tflite/python/lite.py | 3200 | tf.compat.v1.lite.TFLiteConverter.from_keras_model_file( keras_model) | STRING |
| CRITICAL⚡ | tflite/python/lite_v2_test.py | 1906 | tensor_info_x = tf.compat.v1.saved_model.utils.build_tensor_info(x) | CODE |
| CRITICAL⚡ | tflite/python/lite_v2_test.py | 1907 | tensor_info_y = tf.compat.v1.saved_model.utils.build_tensor_info(y) | CODE |
| CRITICAL⚡ | tflite/python/lite_v2_test.py | 1911 | 'serving_default': tf.compat.v1.saved_model.signature_def_utils.build_signature_def( # pylint: disable=line | CODE |
| CRITICAL⚡ | tflite/python/lite_v2_test.py | 1997 | tensor_info_x = tf.compat.v1.saved_model.utils.build_tensor_info(x) | CODE |
| CRITICAL⚡ | tflite/python/lite_v2_test.py | 1998 | tensor_info_y = tf.compat.v1.saved_model.utils.build_tensor_info(y) | CODE |
| CRITICAL⚡ | tflite/python/lite_v2_test.py | 2002 | 'serving_default': tf.compat.v1.saved_model.signature_def_utils.build_signature_def( # pylint: disable=line | CODE |
| CRITICAL | tflite/python/lite_v2_test.py | 1924 | builder = tf.compat.v1.saved_model.builder.SavedModelBuilder( | CODE |
| CRITICAL | tflite/python/lite_v2_test.py | 2014 | builder = tf.compat.v1.saved_model.builder.SavedModelBuilder( | CODE |
| CRITICAL | tflite/testing/zip_test_utils.py | 356 | tensor_info = tf.compat.v1.saved_model.utils.build_tensor_info(tensor) | CODE |
| CRITICAL | tflite/testing/zip_test_utils.py | 598 | tf.compat.v1.saved_model.signature_def_utils.build_signature_def( | CODE |
| CRITICAL | tflite/testing/zip_test_utils.py | 605 | builder = tf.compat.v1.saved_model.builder.SavedModelBuilder( | CODE |
| CRITICAL | tflite/testing/op_tests/dynamic_rnn.py | 49 | lstm_cell = tf.compat.v1.nn.rnn_cell.BasicLSTMCell( | CODE |
| CRITICAL | tflite/testing/op_tests/unfused_gru.py | 39 | cell_fw = tf.compat.v1.nn.rnn_cell.GRUCell(parameters["units"]) | CODE |
| CRITICAL | tflite/testing/op_tests/unfused_gru.py | 40 | cell_bw = tf.compat.v1.nn.rnn_cell.GRUCell(parameters["units"]) | CODE |
| CRITICAL | tflite/testing/op_tests/lstm.py | 57 | lstm_cell = tf.compat.v1.nn.rnn_cell.BasicLSTMCell( | CODE |
| CRITICAL | …te/testing/op_tests/static_rnn_with_control_flow_v2.py | 54 | lstm_cell = tf.compat.v1.nn.rnn_cell.BasicLSTMCell( | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tflite/tools/flatbuffer_utils.py | 0 | converts a numerical enum value to a readable tensor type. | STRING |
| HIGH | tflite/tools/visualize.py | 0 | converts a numerical enum value to a readable tensor type. | STRING |
| HIGH | litert/python/tools/flatbuffer_utils.py | 0 | converts a numerical enum value to a readable tensor type. | STRING |
| HIGH | tflite/tools/pip_package/setup_with_binary.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | ci/setup_with_binary.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | ci/tools/python/vendor_sdk/qualcomm/setup.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | ci/tools/python/vendor_sdk/samsung/setup.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | ci/tools/python/vendor_sdk/mediatek/setup.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | ci/tools/python/vendor_sdk/google_tensor/setup.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | ci/tools/python/wheel/setup_with_binary.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | ci/tools/python/wheel/converter_setup_with_binary.py | 0 | litert is for mobile and embedded devices. litert is the official solution for running machine learning models on mobile | STRING |
| HIGH | …quantizer/tests/end_to_end_tests/concatenation_test.py | 0 | e2e tests for the quantizer for model with transpose. | STRING |
| HIGH | …ert_quantizer/tests/end_to_end_tests/transpose_test.py | 0 | e2e tests for the quantizer for model with transpose. | STRING |
| HIGH | …/litert_quantizer/tests/end_to_end_tests/split_test.py | 0 | e2e tests for the quantizer for model with transpose. | STRING |
| HIGH | …er/litert_quantizer/tests/end_to_end_tests/sum_test.py | 0 | e2e tests for the quantizer for model with transpose. | STRING |
| HIGH | …r_sdk/qualcomm/ai_edge_litert_sdk_qualcomm/__init__.py | 0 | returns the absolute path to the root of the downloaded sdk files. | STRING |
| HIGH | …dor_sdk/samsung/ai_edge_litert_sdk_samsung/__init__.py | 0 | returns the absolute path to the root of the downloaded sdk files. | STRING |
| HIGH | …r_sdk/mediatek/ai_edge_litert_sdk_mediatek/__init__.py | 0 | returns the absolute path to the root of the downloaded sdk files. | STRING |
| HIGH | …le_tensor/ai_edge_litert_sdk_google_tensor/__init__.py | 0 | returns the absolute path to the root of the downloaded sdk files. | STRING |
| HIGH | …b/metadata/python/metadata_writers/image_classifier.py | 0 | creates metadatawriter based on general/input/output information. args: model_buffer: valid buffer of the model file. ge | STRING |
| HIGH | …_lib/metadata/python/metadata_writers/nl_classifier.py | 0 | creates metadatawriter based on general/input/output information. args: model_buffer: valid buffer of the model file. ge | STRING |
| HIGH | …metadata/python/metadata_writers/bert_nl_classifier.py | 0 | creates metadatawriter based on general/input/output information. args: model_buffer: valid buffer of the model file. ge | STRING |
| HIGH | …b/metadata/python/metadata_writers/audio_classifier.py | 0 | creates metadatawriter based on general/input/output information. args: model_buffer: valid buffer of the model file. ge | STRING |
| HIGH | …b/metadata/python/metadata_writers/image_classifier.py | 0 | creates mandatory metadata for litert support inference. the parameters required in this method are mandatory when using | STRING |
| HIGH | …ib/metadata/python/metadata_writers/object_detector.py | 0 | creates mandatory metadata for litert support inference. the parameters required in this method are mandatory when using | STRING |
| HIGH | …ib/metadata/python/metadata_writers/image_segmenter.py | 0 | creates mandatory metadata for litert support inference. the parameters required in this method are mandatory when using | STRING |
| HIGH | litert/python/tools/model_utils/dialect/tfl/_arg_min.py | 0 | returns the axis value if 'dim' is constant, otherwise the ssavalue. | STRING |
| HIGH | …t/python/tools/model_utils/dialect/tfl/_expand_dims.py | 0 | returns the axis value if 'dim' is constant, otherwise the ssavalue. | STRING |
| HIGH | litert/python/tools/model_utils/dialect/tfl/_arg_max.py | 0 | returns the axis value if 'dim' is constant, otherwise the ssavalue. | STRING |
| HIGH | litert/python/aot/vendors/qualcomm/qualcomm_backend.py | 0 | backend implementation for the example compiler plugin.. | STRING |
| HIGH | litert/python/aot/vendors/example/example_backend.py | 0 | backend implementation for the example compiler plugin.. | STRING |
| HIGH | litert/python/aot/vendors/mediatek/mediatek_backend.py | 0 | backend implementation for the example compiler plugin.. | STRING |
| HIGH | litert/python/aot/vendors/qualcomm/qualcomm_backend.py | 0 | backend implementation for the example compiler plugin. | STRING |
| HIGH | litert/python/aot/vendors/example/example_backend.py | 0 | backend implementation for the example compiler plugin. | STRING |
| HIGH | litert/python/aot/vendors/samsung/samsung_backend.py | 0 | backend implementation for the example compiler plugin. | STRING |
| HIGH | litert/python/aot/vendors/mediatek/mediatek_backend.py | 0 | backend implementation for the example compiler plugin. | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | configure.py | 196 | # Check if the path is valid | COMMENT |
| LOW | configure.py | 456 | # Check if current bazel version can be detected properly. | COMMENT |
| LOW | configure.py | 969 | # Check if current clang version can be detected properly. | COMMENT |
| LOW | tflite/tools/build_aar.sh | 59 | # Check if the file exist. | COMMENT |
| LOW | tflite/tools/build_aar.sh | 170 | # Check if users already run configure | COMMENT |
| LOW | tflite/python/util.py | 1135 | # Check if metadata has already been populated. | COMMENT |
| LOW | tflite/python/lite.py | 508 | # Check if representative_dataset is specified. | COMMENT |
| LOW | tflite/python/lite.py | 3338 | # Check if graph is frozen. | COMMENT |
| LOW⚡ | tflite/python/lite_v2_test.py | 1302 | # Check if adding debug mode should output a different flatbuffer. | COMMENT |
| LOW⚡ | tflite/python/lite_v2_test.py | 1305 | # Check if newly added ops are numeric verify ops. | COMMENT |
| LOW | tflite/schema/upgrade_schema.py | 256 | # Check if builtin_code is the appropriate string type | COMMENT |
| LOW | litert_quantizer/litert_quantizer/params_generator.py | 438 | # Check if any of the tensors needs to be duplicated. | COMMENT |
| LOW | litert_quantizer/litert_quantizer/params_generator.py | 444 | # Check if the buffer needs to be duplicated. | COMMENT |
| LOW | …rt_quantizer/litert_quantizer/algorithm_manager_api.py | 175 | # Check if an op quantization config is supported for a given algorithm. | COMMENT |
| LOW | litert_quantizer/litert_quantizer/recipe_test.py | 172 | # Check if the quantized models match. | COMMENT |
| LOW⚡ | litert_quantizer/litert_quantizer/calibrator_test.py | 238 | # Check if internal counter is updated. | COMMENT |
| LOW⚡ | litert_quantizer/litert_quantizer/quantizer_test.py | 435 | # Check if the quant config is supported by default policy. | COMMENT |
| LOW⚡ | litert_quantizer/litert_quantizer/quantizer_test.py | 442 | # Check if the quant config fails on dummy policy. | COMMENT |
| LOW | litert_quantizer/litert_quantizer/quantizer_test.py | 457 | # Check if the quant config is supported by example policy in *.json file. | COMMENT |
| LOW | …tert_quantizer/transformations/quantize_tensor_test.py | 253 | # Check if the scale and zero point tensors are inserted correctly. | COMMENT |
| LOW⚡ | …tert_quantizer/transformations/transformation_utils.py | 139 | # Check if the model has a buffer lookup mapping, and if not, add one. | COMMENT |
| LOW⚡ | …tert_quantizer/transformations/transformation_utils.py | 147 | # Check if we already have a buffer for this data. | COMMENT |
| LOW | …ithms/uniform_quantize/uniform_quantize_tensor_test.py | 391 | # Check if it is safe to cast int64 bias to int32. We save the int32 | COMMENT |
| LOW | …ert_quantizer/algorithms/uniform_quantize/gptq_test.py | 247 | # Check if scales and zp are correctly calculated. | COMMENT |
| LOW | …ert_quantizer/algorithms/uniform_quantize/gptq_test.py | 375 | # Check if scales and zp are correctly calculated. | COMMENT |
| LOW | …antizer/algorithms/uniform_quantize/common_quantize.py | 338 | # Check if SRQ. | COMMENT |
| LOW | …antizer/algorithms/uniform_quantize/common_quantize.py | 385 | # Check if SRQ. | COMMENT |
| LOW | …s/uniform_quantize/op_architecture_tests/test_utils.py | 94 | # Check if SRQ. | COMMENT |
| LOW | …s/uniform_quantize/op_architecture_tests/test_utils.py | 153 | # Check if SRQ. | COMMENT |
| LOW | …s/uniform_quantize/op_architecture_tests/test_utils.py | 387 | # Check if SRQ. | COMMENT |
| LOW | …s/uniform_quantize/op_architecture_tests/test_utils.py | 404 | # Check if WEIGHT_ONLY. | COMMENT |
| LOW | …s/uniform_quantize/op_architecture_tests/test_utils.py | 429 | # Check if SRQ. | COMMENT |
| LOW | …s/uniform_quantize/op_architecture_tests/test_utils.py | 448 | # Check if SRQ. | COMMENT |
| LOW | …rm_quantize/op_architecture_tests/batch_matmul_test.py | 215 | # Check if SRQ. | COMMENT |
| LOW | …rm_quantize/op_architecture_tests/batch_matmul_test.py | 278 | # Check if SRQ. | COMMENT |
| LOW | …quantize/op_architecture_tests/fully_connected_test.py | 111 | # Check if SRQ. | COMMENT |
| LOW | …izer/litert_quantizer/algorithms/utils/common_utils.py | 125 | # Check if find op_config in policy config_check_policy. | COMMENT |
| LOW | …izer/litert_quantizer/algorithms/utils/common_utils.py | 1096 | # Check if SRQ. | COMMENT |
| LOW | …izer/litert_quantizer/algorithms/utils/common_utils.py | 1109 | # Check if DRQ. | COMMENT |
| LOW | …izer/litert_quantizer/algorithms/utils/common_utils.py | 1118 | # Check if WEIGHT_ONLY. | COMMENT |
| LOW⚡ | …litert_quantizer/algorithms/utils/common_utils_test.py | 435 | # Check if DRQ. | COMMENT |
| LOW⚡ | …litert_quantizer/algorithms/utils/common_utils_test.py | 443 | # Check if WEIGHT_ONLY. | COMMENT |
| LOW⚡ | …litert_quantizer/algorithms/utils/common_utils_test.py | 449 | # Check if SRQ. | COMMENT |
| LOW | …litert_quantizer/algorithms/utils/common_utils_test.py | 63 | # Check if WEIGHT_ONLY. | COMMENT |
| LOW | …litert_quantizer/algorithms/utils/common_utils_test.py | 81 | # Check if DRQ. | COMMENT |
| LOW | …litert_quantizer/algorithms/utils/common_utils_test.py | 93 | # Check if SRQ. | COMMENT |
| LOW | …litert_quantizer/algorithms/utils/common_utils_test.py | 419 | # Check if DRQ and CONV_2D. | COMMENT |
| LOW | …antizer/algorithms/nonlinear_quantize/float_casting.py | 66 | # Check if WEIGHT_ONLY. | COMMENT |
| LOW | litert_quantizer/litert_quantizer/tests/mnist_test.py | 275 | # Check if the output model has `integer-only` tensors. | COMMENT |
| LOW⚡ | …r/litert_quantizer/utils/tfl_interpreter_utils_test.py | 401 | # Check if we cover a reasonable range (probability check) | COMMENT |
| LOW | litert/integration_test/device_script_common.sh | 41 | # Print message and exit. | COMMENT |
| LOW | litert/support_lib/metadata/python/metadata.py | 569 | # Check if metadata has already been populated. | COMMENT |
| LOW | litert/support_lib/metadata/python/metadata.py | 624 | # Verify if the number of tensor metadata matches the number of tensors. | COMMENT |
| LOW | …ert/support_lib/metadata/python/tests/metadata_test.py | 274 | # Check if the model buffer read from file is the same as that read from | COMMENT |
| LOW | …ert/support_lib/metadata/python/tests/metadata_test.py | 328 | # Check if the model buffer read from file is the same as that read from | COMMENT |
| LOW | …/python/tests/metadata_writers/metadata_writer_test.py | 178 | # Verify if the metadata populated is correct. | COMMENT |
| LOW | …/python/tests/metadata_writers/metadata_writer_test.py | 185 | # Verify if the associated file is packed as expected. | COMMENT |
| LOW | litert/python/tools/flatbuffer_utils_test.py | 335 | # Check if the pack model's data is updated. | COMMENT |
| LOW | litert/python/tools/mmap_utils.py | 167 | # Write the file in chunks to avoid creating large internal buffers. | COMMENT |
| LOW | litert/python/tools/model_utils/core/utils.py | 108 | # Check if the type is signless (or signed) | COMMENT |
| 5 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …litert/_agents/skills/litert_ats_op_generator/SKILL.md | 57 | ### Step 0: Classify the Op & Verify Prerequisites | COMMENT |
| LOW | …litert/_agents/skills/litert_ats_op_generator/SKILL.md | 144 | ### Step 1: Author the Generator (`test/generators/<op>.h` & `test/generators/BUILD`) | COMMENT |
| LOW | …litert/_agents/skills/litert_ats_op_generator/SKILL.md | 249 | ### Step 2: Export in `generators.h` & `test/generators/BUILD` | COMMENT |
| LOW | …litert/_agents/skills/litert_ats_op_generator/SKILL.md | 261 | ### Step 3: Register the Op in ATS (`litert/ats/`) | COMMENT |
| LOW | …litert/_agents/skills/litert_ats_op_generator/SKILL.md | 299 | ### Step 4: Run Mandatory Verification Gates & Configure Backend Exclusions | COMMENT |
| LOW | tflite/CMakeLists.txt | 19 | # WARNING: This is an experimental that is subject to change. | COMMENT |
| LOW | tflite/kernels/internal/runtime_shape.h | 204 | // NOTE: This function does not handle potential integer overflow. Use | COMMENT |
| LOW | tflite/core/subgraph.h | 561 | // NOTE: This function is expected to be paired with ReleaseSubgraphContext() | COMMENT |
| LOW | tflite/core/subgraph.h | 574 | // NOTE: This function is expected to be used after AcquireSubgraphContext() | COMMENT |
| LOW | tflite/core/subgraph.h | 591 | // NOTE: This function is expected to be called only when the subgraph that | COMMENT |
| LOW | tflite/core/subgraph.h | 622 | // NOTE: This function is expected to be called only when this subgraph will | COMMENT |
| LOW | tflite/core/interpreter.h | 241 | /// NOTE: This function does not affect cancellation triggered by the callback | COMMENT |
| LOW | tflite/core/async/c/task.h | 60 | /// NOTE: This method does not check tensor index is pointing to a valid tensor. | COMMENT |
| LOW | tflite/core/async/c/task.h | 118 | /// NOTE: This method does not check tensor index is pointing to a | COMMENT |
| LOW | tflite/core/c/c_api.h | 542 | /// NOTE: This function is similar to TfLiteInterpreterResizeInputTensorStrict() | COMMENT |
| LOW | tflite/core/c/c_api.h | 547 | /// NOTE: This function makes a copy of the input dimensions, so the caller can | COMMENT |
| LOW | tflite/core/c/c_api_opaque.h | 559 | /// NOTE: This function is expected to be paired with | COMMENT |
| LOW | tflite/core/c/c_api_opaque.h | 572 | /// NOTE: This function is expected to be used after | COMMENT |
| LOW | tflite/core/c/c_api_opaque.h | 590 | /// NOTE: This function is expected to be called only when the subgraph that | COMMENT |
| LOW⚡ | tflite/core/c/common.h | 1573 | // NOTE: This function is part of the TensorFlow Lite Extension APIs, see above. | COMMENT |
| LOW⚡ | tflite/core/c/common.h | 1579 | // NOTE: This function is part of the TensorFlow Lite Extension APIs, see above. | COMMENT |
| LOW⚡ | tflite/core/c/common.h | 1585 | // NOTE: This function is part of the TensorFlow Lite Extension APIs, see above. | COMMENT |
| LOW | tflite/core/c/common.h | 1134 | /// NOTE: This function is expected to be paired with ReleaseSubgraphContext() | COMMENT |
| LOW | tflite/core/c/common.h | 1145 | /// NOTE: This function is expected to be used after AcquireSubgraphContext() | COMMENT |
| LOW | tflite/converter/kernels/internal/runtime_shape.h | 202 | // NOTE: This function does not handle potential integer overflow. | COMMENT |
| LOW | tflite/python/convert.py | 177 | # WARNING: These ops are currently experimental and have not yet been | COMMENT |
| LOW | tflite/delegates/utils.h | 47 | // NOTE: This function is expected to be paired with ReleaseSubgraphContext() | COMMENT |
| LOW | tflite/delegates/utils.h | 56 | // NOTE: This function is expected to be used after AcquireSubgraphContext() | COMMENT |
| LOW | tflite/delegates/utils.h | 70 | // NOTE: This function is expected to be called only when the subgraph that | COMMENT |
| LOW | …le_delegate/sample_stable_delegate_with_control_flow.h | 70 | // NOTE: This function is expected to be called recursively to gather all the | COMMENT |
| LOW | litert_quantizer/README.md | 214 | ### Step 1: Initialize Quantizer with Source Model | COMMENT |
| LOW | litert_quantizer/README.md | 225 | ### Step 2: Choose & Load a Quantization Recipe | COMMENT |
| LOW | litert_quantizer/README.md | 267 | ### Step 3: Calibrate with Representative Data (Static Quantization Only) | COMMENT |
| LOW | litert_quantizer/README.md | 353 | ### Step 4: Quantize & Export the Model | COMMENT |
| LOW | litert_quantizer/README.md | 366 | ### Step 5: Validate Numerical Accuracy | COMMENT |
| LOW | litert_quantizer/README.md | 401 | ### Step 6: Visualize Models with Model Explorer | COMMENT |
| LOW | litert_quantizer/README.md | 414 | ### Step 7: Deploy on Edge Hardware | COMMENT |
| LOW | …tert_quantizer/experimental/skills/auto_quant/SKILL.md | 94 | ### Step 1: Context & Role Declaration | COMMENT |
| LOW | …tert_quantizer/experimental/skills/auto_quant/SKILL.md | 105 | ### Step 2: Technical Design & File List | COMMENT |
| LOW | …tert_quantizer/experimental/skills/auto_quant/SKILL.md | 113 | ### Step 3: Experiment Harness Architecture | COMMENT |
| LOW | …tert_quantizer/experimental/skills/auto_quant/SKILL.md | 130 | ### Step 4: Verification Plan | COMMENT |
| LOW | …tert_quantizer/experimental/skills/auto_quant/SKILL.md | 140 | ### Step 5: Post-Execution Walkthrough | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 33 | ### Step 1: Install Dependencies | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 57 | ### Step 2: Clone and Build | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 71 | ### Step 3: Verify Installation | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 82 | ### Step 1: Install Prerequisites | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 95 | ### Step 2: Build for macOS | COMMENT |
| LOW⚡ | docs/instructions/BUILD_INSTRUCTIONS.md | 115 | ### Step 1: Install Prerequisites | COMMENT |
| LOW⚡ | docs/instructions/BUILD_INSTRUCTIONS.md | 121 | ### Step 2: Configure Environment | COMMENT |
| LOW⚡ | docs/instructions/BUILD_INSTRUCTIONS.md | 131 | ### Step 3: Build for Windows | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 161 | ### Step 1: Install Android Development Tools | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 179 | ### Step 2: Build for Android with Bazel | COMMENT |
| LOW | docs/instructions/BUILD_INSTRUCTIONS.md | 201 | ### Step 3: Test Accelerator Support for Android | COMMENT |
| LOW | docs/instructions/GPU_BUILD_INSTRUCTIONS.md | 19 | ## Step 1: Clone the LiteRT Repository | COMMENT |
| LOW | docs/instructions/GPU_BUILD_INSTRUCTIONS.md | 33 | ## Step 2: Build the Android GPU Accelerator | COMMENT |
| LOW | docs/instructions/GPU_BUILD_INSTRUCTIONS.md | 47 | ## Step 3: Locate Build Artifacts & Deploy to Device | COMMENT |
| LOW | tensor/backends/testing/README.md | 30 | ### Step 1: Implement `TestBackendBridge` | COMMENT |
| LOW | tensor/backends/testing/README.md | 83 | ### Step 2: Define Backend Traits | COMMENT |
| LOW | tensor/backends/testing/README.md | 97 | ### Step 3: Instantiate the Suite | COMMENT |
| LOW | tensor/backends/testing/README.md | 113 | ### Step 4: Update the BUILD target | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | configure.py | 260 | CODE | |
| LOW | configure.py | 275 | CODE | |
| LOW | configure.py | 1021 | CODE | |
| LOW | configure.py | 1128 | CODE | |
| LOW | configure.py | 1262 | CODE | |
| LOW | tflite/tools/flatbuffer_utils.py | 358 | CODE | |
| LOW | tflite/tools/randomize_weights.py | 49 | CODE | |
| LOW | tflite/tools/zip_files.py | 27 | CODE | |
| LOW | tflite/tools/visualize.py | 318 | CODE | |
| LOW | tflite/tools/visualize.py | 422 | CODE | |
| LOW | tflite/tools/optimize/debugging/python/debugger.py | 332 | CODE | |
| LOW | tflite/python/util.py | 173 | CODE | |
| LOW | tflite/python/util.py | 332 | CODE | |
| LOW | tflite/python/util.py | 649 | CODE | |
| LOW | tflite/python/util.py | 756 | CODE | |
| LOW | tflite/python/util.py | 876 | CODE | |
| LOW | tflite/python/util.py | 1013 | CODE | |
| LOW | tflite/python/util.py | 1116 | CODE | |
| LOW | tflite/python/util.py | 343 | CODE | |
| LOW | tflite/python/op_hint.py | 828 | CODE | |
| LOW | tflite/python/op_hint.py | 932 | CODE | |
| LOW | tflite/python/op_hint.py | 1059 | CODE | |
| LOW | tflite/python/op_hint.py | 1168 | CODE | |
| LOW | tflite/python/op_hint.py | 558 | CODE | |
| LOW | tflite/python/convert.py | 915 | CODE | |
| LOW | tflite/python/convert.py | 1071 | CODE | |
| LOW | tflite/python/util_test.py | 301 | CODE | |
| LOW | tflite/python/lite.py | 596 | CODE | |
| LOW | tflite/python/lite.py | 1355 | CODE | |
| LOW | tflite/python/lite.py | 1649 | CODE | |
| LOW | tflite/python/lite.py | 2550 | CODE | |
| LOW | tflite/python/lite.py | 3271 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 3580 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 4875 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 4876 | CODE | |
| LOW | tflite/python/tflite_keras_util.py | 55 | CODE | |
| LOW | tflite/python/optimize/calibrator.py | 97 | CODE | |
| LOW | tflite/testing/zip_test_utils.py | 114 | CODE | |
| LOW | tflite/testing/zip_test_utils.py | 145 | CODE | |
| LOW | tflite/testing/zip_test_utils.py | 259 | CODE | |
| LOW | tflite/testing/zip_test_utils.py | 368 | CODE | |
| LOW | tflite/testing/mlir_convert.py | 30 | CODE | |
| LOW | tflite/testing/mlir_convert.py | 143 | CODE | |
| LOW | tflite/schema/upgrade_schema.py | 147 | CODE | |
| LOW | tflite/schema/upgrade_schema.py | 265 | CODE | |
| LOW | litert_quantizer/litert_quantizer/recipe_manager.py | 157 | CODE | |
| LOW | …quantizer/litert_quantizer/transformation_performer.py | 148 | CODE | |
| LOW | litert_quantizer/litert_quantizer/params_generator.py | 69 | CODE | |
| LOW | litert_quantizer/litert_quantizer/params_generator.py | 210 | CODE | |
| LOW | litert_quantizer/litert_quantizer/calibrator.py | 501 | CODE | |
| LOW | litert_quantizer/litert_quantizer/model_validator.py | 280 | CODE | |
| LOW | litert_quantizer/litert_quantizer/qtyping.py | 447 | CODE | |
| LOW | litert_quantizer/litert_quantizer/model_modifier.py | 219 | CODE | |
| LOW | …tert_quantizer/transformation_instruction_generator.py | 247 | CODE | |
| LOW | …tert_quantizer/transformation_instruction_generator.py | 407 | CODE | |
| LOW | …tert_quantizer/transformation_instruction_generator.py | 701 | CODE | |
| LOW | litert_quantizer/litert_quantizer/default_policy.py | 326 | CODE | |
| LOW | …rt_quantizer/litert_quantizer/params_generator_test.py | 608 | CODE | |
| LOW | …tert_quantizer/transformations/transformation_utils.py | 80 | CODE | |
| LOW | …algorithms/uniform_quantize/uniform_quantize_tensor.py | 48 | CODE | |
| 44 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tflite/types/half.h | 24 | // known to be at least a robust f16<->f32 conversion, which generally means a | COMMENT |
| MEDIUM | tflite/python/convert.py | 780 | # leverage defaults specified in proto definition. | COMMENT |
| MEDIUM | tflite/python/lite_v2_test.py | 596 | # the bias tensor should be 32 bits to utilize optimized kernels, | COMMENT |
| MEDIUM | …delegates/gpu/common/tasks/mean_stddev_normalization.h | 68 | // provide more stable and robust results | COMMENT |
| MEDIUM | …delegates/gpu/common/tasks/mean_stddev_normalization.h | 133 | // provide more stable and robust results | COMMENT |
| MEDIUM | litert_quantizer/litert_quantizer/calibrator_test.py | 606 | # operator and rank of the tensor. We flatten them here for robust | COMMENT |
| LOW | …tert_quantizer/transformations/transformation_utils.py | 100 | # If the model already has the op code, just return the index. | COMMENT |
| LOW | ci/run_bazel_build.sh | 240 | # buildable. To avoid creating new TAP targets, simply add benchmark | COMMENT |
| LOW | ci/tools/python/vendor_sdk/google_tensor/setup.py | 106 | # If we simply return None, the build process will continue, which is | COMMENT |
| LOW⚡ | ci/tools/python/vendor_sdk/google_tensor/setup.py | 213 | # If we simply return None, the build process will continue, which | COMMENT |
| LOW⚡ | ci/tools/python/vendor_sdk/google_tensor/setup.py | 229 | # If we simply return None, the build process will continue, which is not | COMMENT |
| MEDIUM | …/vendors/intel_openvino/dispatch/weight_bank_runtime.h | 59 | // are matched by the Parameter's friendly_name, so binding is robust to input | COMMENT |
| MEDIUM | …endors/intel_openvino/compiler/weights_to_parameters.h | 36 | // its pool buffer by friendly_name (robust to input reordering across import). | COMMENT |
| MEDIUM | litert/vendors/intel_openvino/compiler/global_graph.h | 38 | // at dispatch (matched by name so binding is robust to input reordering across | COMMENT |
| MEDIUM | …ndors/intel_openvino/compiler/alias_shared_constants.h | 46 | // robust guard -- no need to enumerate which transforms alter content. | COMMENT |
| MEDIUM | litert/vendors/google_tensor/dispatch/sb_api.h | 91 | // implementation to utilize a proprietary fence, which generally have superior | COMMENT |
| MEDIUM | litert/core/dispatch_op_schema.h | 21 | // functions leverage the flexbuffer api under the hood which allows for inplace | COMMENT |
| LOW⚡ | litert/python/tools/mmap_utils_test.py | 242 | # This should not raise any exception and should just return. | COMMENT |
| LOW | litert/python/tools/mmap_utils_test.py | 306 | # This should not raise any exception and should just return. | COMMENT |
| LOW | litert/python/aot/ai_pack/export_lib.py | 164 | # Skip non-MTK targets, just create a placeholder file. | COMMENT |
| MEDIUM | litert/cc/internal/litert_c_types_printing.h | 33 | /// @todo Migrate the code in `tools/dump.h` to leverage the Abseil stringify | COMMENT |
| MEDIUM | ml_drift_delegate/delegate/delegate_options.h | 215 | // If true, the weights rearrangement, if supported, will utilize GPU. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tflite/python/convert.py | 146 | except UnicodeDecodeError: | CODE |
| HIGH | tflite/python/interpreter.py | 734 | except ValueError: | CODE |
| HIGH | tflite/python/lite_v2_test.py | 5311 | except ValueError: | CODE |
| HIGH | …/vendor_sdk/intel/ai_edge_litert_sdk_intel/__init__.py | 110 | except OSError: | CODE |
| HIGH | …/vendor_sdk/intel/ai_edge_litert_sdk_intel/__init__.py | 125 | except OSError: | CODE |
| HIGH | litert/python/tools/benchmark_litert_model.py | 86 | except ImportError: | CODE |
| HIGH | …n/aot/vendors/intel_openvino/intel_openvino_backend.py | 51 | except FileNotFoundError: | CODE |
| HIGH | …n/aot/vendors/intel_openvino/intel_openvino_backend.py | 63 | except (ModuleNotFoundError, TypeError): | CODE |
| HIGH | …n/aot/vendors/intel_openvino/intel_openvino_backend.py | 190 | except ImportError: | CODE |
| HIGH | litert/python/aot/core/apply_plugin.py | 170 | except OSError: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | configure.py | 1286 | print('Error retrieving bazel version: ', e.output.decode('UTF-8').strip()) | CODE |
| LOW | tflite/python/util.py | 1151 | except Exception: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/convert.py | 386 | except Exception as e: | CODE |
| LOW | tflite/python/convert.py | 437 | except Exception as e: | CODE |
| LOW | tflite/python/convert_phase.py | 213 | except Exception as error: | CODE |
| MEDIUM | tflite/python/convert_phase.py | 203 | def wrapper(*args, **kwargs): | CODE |
| LOW | tflite/python/lite.py | 1202 | except Exception: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/lite.py | 1714 | except Exception as error: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/lite.py | 1794 | except Exception as e: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/lite.py | 1957 | except Exception: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/lite.py | 2120 | except Exception: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/lite.py | 2639 | except Exception: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/lite.py | 2974 | except Exception: # pylint: disable=broad-except | CODE |
| LOW | tflite/python/optimize/calibrator.py | 80 | except Exception as e: | CODE |
| LOW | tflite/testing/mlir_convert.py | 137 | except Exception as e: # pylint: disable=broad-except | CODE |
| LOW | …izer/litert_quantizer/algorithms/utils/common_utils.py | 288 | except Exception as e: | CODE |
| MEDIUM | ci/tools/python/vendor_sdk/sdist_wrapper.py | 174 | print("Error running setup.py sdist:", file=sys.stderr) | CODE |
| LOW | ci/tools/python/vendor_sdk/intel/setup.py | 303 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | ci/tools/python/vendor_sdk/intel/setup.py | 326 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | ci/tools/python/vendor_sdk/qualcomm/setup.py | 85 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | ci/tools/python/vendor_sdk/qualcomm/setup.py | 186 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | …r_sdk/qualcomm/ai_edge_litert_sdk_qualcomm/__init__.py | 57 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| MEDIUM | …r_sdk/qualcomm/ai_edge_litert_sdk_qualcomm/__init__.py | 58 | print(f"Error determining SDK path: {e}", file=sys.stderr) | CODE |
| LOW | ci/tools/python/vendor_sdk/samsung/setup.py | 73 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | ci/tools/python/vendor_sdk/samsung/setup.py | 174 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | …dor_sdk/samsung/ai_edge_litert_sdk_samsung/__init__.py | 59 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| MEDIUM | …dor_sdk/samsung/ai_edge_litert_sdk_samsung/__init__.py | 60 | print(f"Error determining SDK path: {e}", file=sys.stderr) | CODE |
| LOW | ci/tools/python/vendor_sdk/mediatek/setup.py | 71 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | ci/tools/python/vendor_sdk/mediatek/setup.py | 176 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | …r_sdk/mediatek/ai_edge_litert_sdk_mediatek/__init__.py | 78 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| MEDIUM | …r_sdk/mediatek/ai_edge_litert_sdk_mediatek/__init__.py | 79 | print(f"Error determining SDK path: {e}", file=sys.stderr) | CODE |
| LOW | ci/tools/python/vendor_sdk/google_tensor/setup.py | 99 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW⚡ | ci/tools/python/vendor_sdk/google_tensor/setup.py | 223 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | …le_tensor/ai_edge_litert_sdk_google_tensor/__init__.py | 61 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| MEDIUM | …le_tensor/ai_edge_litert_sdk_google_tensor/__init__.py | 62 | print(f"Error determining SDK path: {e!r}", file=sys.stderr) | CODE |
| LOW | ci/tools/python/wheel/setup_with_binary.py | 80 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | ci/tools/python/wheel/setup_with_binary.py | 131 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | ci/tools/python/wheel/converter_setup_with_binary.py | 71 | except Exception as e: # pylint: disable=broad-exception-caught | CODE |
| LOW | litert/vendors/nvidia/run_head.sh | 285 | except Exception as exc: errors.append(str(exc)) | CODE |
| MEDIUM | litert/vendors/nvidia/run_head.sh | 266 | def sample(): | CODE |
| LOW | litert/python/litert_wrapper/simple_accel_smoke_test.py | 192 | except Exception as exc: | CODE |
| LOW | .github/workflows/clang_tidy.yml | 138 | except Exception: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | litert/vendors/broadcom/example/CMakeLists.txt | 2 | # Portions generated by AI | COMMENT |
| HIGH | litert/vendors/broadcom/dispatch/CMakeLists.txt | 2 | # Portions generated by AI | COMMENT |
| HIGH | litert/vendors/broadcom/common/CMakeLists.txt | 2 | # Portions generated by AI | COMMENT |
| HIGH | litert/vendors/broadcom/compiler/CMakeLists.txt | 2 | # Portions generated by AI | COMMENT |
| HIGH | litert/python/tools/model_utils/dialect/tfl/_fill.py | 102 | # Do not check ndim as requested | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …tal/microfrontend/python/ops/audio_microfrontend_op.py | 28 | CODE | |
| LOW | tflite/tools/optimize/debugging/python/debugger_test.py | 46 | CODE | |
| LOW | tflite/python/convert.py | 263 | CODE | |
| LOW | tflite/python/convert.py | 490 | CODE | |
| LOW | tflite/python/convert.py | 909 | CODE | |
| LOW | tflite/python/convert.py | 997 | CODE | |
| LOW | tflite/python/convert.py | 1013 | CODE | |
| LOW | tflite/python/convert.py | 1040 | CODE | |
| LOW | tflite/python/interpreter.py | 402 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 293 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 919 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 1425 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 5341 | CODE | |
| LOW | tflite/python/lite_v2_test.py | 5379 | CODE | |
| LOW | tflite/python/metrics/metrics_nonportable_test.py | 192 | CODE | |
| LOW | tflite/testing/op_tests/conv_activation.py | 146 | CODE | |
| LOW | tflite/testing/op_tests/conv_bias_activation.py | 73 | CODE | |
| LOW | tflite/testing/op_tests/conv_bias_activation.py | 78 | CODE | |
| LOW | litert_quantizer/litert_quantizer/tests/mnist_test.py | 51 | CODE | |
| LOW | …quantizer/tests/end_to_end_tests/strided_slice_test.py | 45 | CODE | |
| LOW | …/litert_quantizer/tests/end_to_end_tests/rsqrt_test.py | 47 | CODE | |
| LOW | …quantizer/tests/end_to_end_tests/concatenation_test.py | 45 | CODE | |
| LOW | …er/litert_quantizer/tests/end_to_end_tests/add_test.py | 52 | CODE | |
| LOW | …tert_quantizer/tests/end_to_end_tests/logistic_test.py | 43 | CODE | |
| LOW | …/litert_quantizer/tests/end_to_end_tests/split_test.py | 45 | CODE | |
| LOW | …_quantizer/tests/end_to_end_tests/input_output_test.py | 49 | CODE | |
| LOW | …r/litert_quantizer/tests/end_to_end_tests/mean_test.py | 45 | CODE | |
| LOW | …ntizer/tests/end_to_end_tests/conv2d_transpose_test.py | 44 | CODE | |
| LOW | …on/tools/model_utils/dialect/tfl/_depthwise_conv_2d.py | 50 | CODE | |
| LOW | litert/python/tools/model_utils/dialect/tfl/_conv_2d.py | 50 | CODE | |
| LOW | tensor/examples/gemma4/helpers/reference/attention.py | 42 | CODE | |
| LOW | …nsor/examples/gemma4/helpers/reference/gemma4_graph.py | 81 | CODE | |
| LOW | tensor/examples/gemma4/helpers/reference/transformer.py | 29 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …imental/acceleration/compatibility/gpu_compatibility.h | 39 | // Example usage: | COMMENT |
| LOW⚡ | tflite/testing/zip_test_utils.py | 47 | # Usage: | COMMENT |
| LOW | tflite/delegates/gpu/api.h | 19 | // Usage example: | COMMENT |
| LOW | tflite/delegates/gpu/common/task/util.h | 60 | // Example usage: | COMMENT |
| LOW | tflite/delegates/gpu/cl/api.h | 33 | // Usage example: | COMMENT |
| LOW | litert/rust/example/segmentation_async_main.rs | 19 | /// Example usage: | COMMENT |
| LOW | litert/rust/example/segmentation_main.rs | 19 | /// Example usage: | COMMENT |
| LOW | litert/rust/src/environment.rs | 63 | /// Example usage: | COMMENT |
| LOW | litert/tflite_support/op_resolver/op_resolver.h | 31 | /// Example usage: | COMMENT |
| LOW | litert/tflite_support/custom_op/custom_op.h | 29 | /// Example usage: | COMMENT |
| LOW | litert/tflite_support/custom_op/custom_op.h | 44 | /// Example usage: | COMMENT |
| LOW | litert/cc/litert_options.h | 340 | /// Example usage: | COMMENT |
| LOW | litert/c/options/litert_runtime_options.h | 25 | // Example usage: | COMMENT |
| LOW | docker_build/build_wheel_with_docker.sh | 21 | # Usage: | COMMENT |
| LOW | docker_build/setup_bazel_env.sh | 21 | # Usage: | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | tflite/testing/op_tests/mirror_pad.py | 46 | "type": ["placeholder"], | CODE |
| LOW⚡ | tflite/testing/op_tests/mirror_pad.py | 53 | "type": ["placeholder"], | CODE |
| LOW⚡ | tflite/testing/op_tests/mirror_pad.py | 60 | "type": ["placeholder"], | CODE |
| LOW | tflite/testing/op_tests/mirror_pad.py | 82 | "type": ["placeholder"], | CODE |
| LOW⚡ | …ntizer/litert_quantizer/utils/constrained_ops_utils.py | 54 | dummy_value = [qtyping.TensorTransformationParams("")] * 2 | CODE |
| LOW⚡ | …ntizer/litert_quantizer/utils/constrained_ops_utils.py | 55 | return dummy_value | CODE |
| LOW | litert/python/tools/mmap_utils_test.py | 52 | b'Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod' | CODE |
| LOW | litert/python/tools/mmap_utils_test.py | 52 | b'Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod' | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tflite/tools/visualize.py | 158 | d.fx = d.fy = null; | CODE |
| HIGH | …t_quantizer/litert_quantizer/utils/validation_utils.py | 166 | KL(data2 || data1) = sum(data2 * log(data2 / data1)). | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | litert/python/tools/model_utils/model_builder.py | 25 | __all__ = [ | CODE |
| LOW | …/python/tools/model_utils/core/_op_building_context.py | 23 | __all__ = ["OpBuildingContext"] | CODE |
| LOW | litert/python/tools/model_utils/core/dialect_base.py | 27 | __all__ = [ | CODE |
| LOW | litert/python/tools/model_utils/core/pass_base.py | 24 | __all__ = ["ModulePassBase", "RewritePatternPassBase"] | CODE |
| LOW | litert/python/tools/model_utils/core/utils.py | 21 | __all__ = [ | CODE |
| LOW | …ython/tools/model_utils/match/_match_dag/tblgen_def.py | 25 | __all__ = ["registry"] | CODE |
| LOW | litert/python/mlir/__init__.py | 20 | __all__ = [] | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tflite/python/lite.py | 1595 | # We make sure to clear the saved_model_dir as there is some | COMMENT |
| LOW | litert/c/internal/litert_runtime_c_api.h | 42 | // don't forget to also bump the ABI version number. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | litert/vendors/qualcomm/README.md | 93 | 1. [Bring state-of-the-art agentic skills to the edge with Gemma 4](https://developers.googleblog.com/bring-state-o | CODE |