tensorflow
모두를 위한 오픈소스 머신러닝 프레임워크
파일 탐색기
최종 버전 다운로드 (.zip)파일 수가 많아 일부만 표시됩니다. 전체 파일은 위 다운로드 버튼으로 확인해 주세요.
- styleguide.md
- config.yml
- tensorflow_issue_template.yaml
- tflite-converter-issue.md
- tflite-in-play-services.md
- tflite-op-request.md
- tflite-other.md
- arm-cd.yml
- arm-ci-extended-cpp.yml
- arm-ci-extended.yml
- arm-ci.yml
- build-reusable.yml
- cffconvert.yml
- check_license.yml
- ci-build.yml
- create_issue.js
- gemini.yml
- issue-on-pr-rollback.yml
- osv-scanner-scheduled.yml
- pylint-presubmit.yml
- release-branch-cherrypick.yml
- scorecards-analysis.yml
- stale-issues.yml
- update-nightly.yml
- update-rbe.yml
- bot_config.yml
- dependabot.yml
- Dockerfile
- Dockerfile
- README.md
- build_devtoolset.sh
- fixlinks.sh
- glibc2.17-inline.patch
- rpm-patch.sh
- builder.packages.txt
- builder.requirements.txt
- cmake_toolchain.packages.txt
- cuda12.1_cudnn9.1.packages.txt
- cuda12.1_cudnn9.8.packages.txt
- cuda12.3_cudnn9.1.packages.txt
- cuda12.8_cudnn9.8.packages.txt
- cuda13.0_cudnn9.12.packages.txt
- cuda13.0_cudnn9.15.packages.txt
- cuda13.2_cudnn9.15.packages.txt
- Dockerfile
- rbe_nvidia.packages.txt
- README.md
- setup.packages.sh
- setup.python.sh
- setup.sources.cudnn.sh
- setup.sources.sh
- setup.toolchain.cmake.sh
- bzlmod
- ci_default
- disk_cache
- enable_pycpp_build
- linux_arm64
- linux_arm64_cross_compile
- linux_arm64_onednn
- linux_x86
- linux_x86_cuda
- linux_x86_cuda13_nvcc
- macos_arm64
- nightly_upload
- no_docker
- no_upload
- numpy1
- public_cache
- public_cache_push
- py310
- py311
- py312
- py313
- py314
- rbe
- versions_upload
- windows_x86_2022
- windows_x86_2022_ml_actions
- windows_x86_ml_actions
- requirements.in
- requirements_lock_3_10.txt
- requirements_lock_3_11.txt
- requirements_lock_3_12.txt
- BUILD.bazel
- nvidia-requirements.txt
- README.md
- requirements.in
- cleanup_docker.sh
- cleanup_summary.sh
- code_check_changed_files.bats
- code_check_full.bats
- convert_msys_paths_to_win_paths.py
- extract_resultstore_links.py
- generate_index_html.sh
- rename_and_verify_wheels.sh
- setup.sh
- setup_docker.sh
- setup_macos.sh
- windows.sh
- any.sh
- bisect.sh
- code_check_changed_files.sh
- code_check_full.sh
- debug_tfci.sh
- pycpp.sh
- README.md
- upload.sh
- wheel.sh
- README.md
- BUILD
- parallel_device.cc
- parallel_device.h
- parallel_device_lib.cc
- parallel_device_lib.h
- parallel_device_lib_test.cc
- parallel_device_remote_test.cc
- parallel_device_test.cc
- parallel_device_testlib.cc
- parallel_device_testlib.h
- abstract_context.h
- abstract_function.h
- abstract_op_attrs.h
- abstract_operation.h
- abstract_tensor_handle.cc
- abstract_tensor_handle.h
- BUILD
- c_api.cc
- c_api.h
- c_api_cluster_test.cc
- c_api_debug.cc
- c_api_debug_test.cc
- c_api_distributed_test.cc
- c_api_experimental.cc
- c_api_experimental.h
- c_api_experimental_reader.cc
- c_api_experimental_reader.h
- c_api_experimental_reader_test.cc
- c_api_experimental_test.cc
- c_api_internal.h
- c_api_remote_function_test.cc
- c_api_remote_test.cc
- c_api_remote_test_util.cc
- c_api_remote_test_util.h
- c_api_test.cc
- c_api_test_util.cc
- c_api_test_util.h
- c_api_unified_experimental.cc
- c_api_unified_experimental.h
- c_api_unified_experimental_eager.cc
- c_api_unified_experimental_graph.cc
- c_api_unified_experimental_internal.h
- c_api_unified_experimental_test.cc
- custom_device_test.cc
- custom_device_testutil.cc
- custom_device_testutil.h
- dlpack.cc
- dlpack.h
- dlpack_test.cc
- gradient_checker.cc
- gradient_checker.h
- gradient_checker_test.cc
- gradients.cc
- gradients.h
- gradients_internal.h
- gradients_test.cc
- graph_function.cc
- graph_function.h
- immediate_execution_context.h
- immediate_execution_distributed_manager.h
- immediate_execution_operation.h
- immediate_execution_tensor_handle.cc
- immediate_execution_tensor_handle.h
- tape.h
- tfe_cancellation_manager_internal.h
- tfe_context_internal.h
- tfe_executor_internal.h
- tfe_monitoring_internal.h
- tfe_monitoring_reader_internal.h
- tfe_op_attrs_internal.h
- tfe_op_internal.h
- tfe_tensor_debug_info_internal.h
- tfe_tensorhandle_internal.h
- tracing_utils.cc
- tracing_utils.h
- unified_api_test.cc
- unified_api_testutil.cc
- unified_api_testutil.h
- BUILD
- cleanup.h
- expiring_lru_cache.h
- expiring_lru_cache_test.cc
- gcs_filesystem.cc
- gcs_filesystem.h
- gcs_filesystem_test.cc
- gcs_helper.cc
- gcs_helper.h
- ram_file_block_cache.cc
- ram_file_block_cache.h
- ram_file_block_cache_test.cc
- BUILD
- copy_file.h
- copy_file_linux.cc
- copy_file_portable.cc
- posix_filesystem.cc
- posix_filesystem.h
- posix_filesystem_helper.cc
- posix_filesystem_helper.h
- posix_filesystem_static.cc
- BUILD
- windows_filesystem.cc
- BUILD
- filesystem_interface.h
- modular_filesystem.cc
- modular_filesystem.h
- modular_filesystem_registration.cc
- modular_filesystem_registration.h
- modular_filesystem_test.cc
- BUILD
- tape_context.cc
- tape_context.h
- tape_operation.cc
- tape_operation.h
- array_grad.cc
- array_grad.h
- array_grad_test.cc
- BUILD
- custom_gradient_test.cc
- grad_test_helper.cc
- grad_test_helper.h
- math_grad.cc
- math_grad.h
- math_grad_test.cc
- nn_grad.cc
- nn_grad.h
- nn_grad_test.cc
- not_differentiable.cc
- not_differentiable.h
- BUILD
- grappler.cc
- grappler.h
- grappler_internal.h
- grappler_test.cc
- BUILD
- c_api.cc
- c_api.h
- tensor_pjrt_buffer_util.cc
- tensor_pjrt_buffer_util.h
- tensor_pjrt_buffer_util_test.cc
- BUILD
- case_format.cc
- case_format.h
- case_format_test.cc
- controller.cc
- controller.h
- path_config.cc
- path_config.h
- source_code.cc
- source_code.h
- view_util.cc
- view_util.h
- BUILD
- testing_ops.cc.golden
- testing_ops.h.golden
- BUILD
- cpp_config.cc
- cpp_config.h
- cpp_file_renderer.cc
- cpp_file_renderer.h
- guard_renderer.cc
- guard_renderer.h
- include_renderer.cc
- include_renderer.h
- namespace_renderer.cc
- namespace_renderer.h
- op_comment_renderer.cc
- op_comment_renderer.h
- op_implementation_renderer.cc
- op_implementation_renderer.h
- op_renderer.cc
- op_renderer.h
- renderer.cc
- renderer.h
- renderer_context.h
- renderer_test.cc
- arg_type_view.cc
- arg_type_view.h
- arg_view.cc
- arg_view.h
- attr_view.cc
- attr_view.h
- BUILD
- op_argument_view.cc
- op_argument_view.h
- op_view.cc
- op_view.h
- BUILD
- cpp_generator.cc
- cpp_generator.h
- cpp_generator_test.cc
- arg_spec.cc
- arg_spec.h
- arg_type.cc
- arg_type.h
- attr_spec.cc
- attr_spec.h
- BUILD
- op_spec.cc
- op_spec.h
- BUILD
- generate_cpp_main.cc
- README.md
- array_ops.cc
- array_ops.h
- BUILD
- io_ops.cc
- io_ops.h
- math_ops.cc
- math_ops.h
- nn_ops.cc
- nn_ops.h
- README.md
- resource_variable_ops.cc
- resource_variable_ops.h
- update_cpp_ops.sh
- BUILD
- pluggable_profiler.cc
- pluggable_profiler.h
- pluggable_profiler_internal.h
- BUILD
- restore_ops.cc
- restore_ops.h
- restore_ops_test.cc
- variable_ops.cc
- variable_ops.h
- variable_ops_test.cc
- asset.cc
- asset.h
- BUILD
- constant.cc
- constant.h
- flat_tensor_function.cc
- flat_tensor_function.h
- partially_revived_objects.cc
- partially_revived_objects.h
- restored_resource.cc
- restored_resource.h
- restored_resource_revival_state.h
- revived_objects.h
- tensorhandle_convertible.h
- tf_concrete_function.cc
- tf_concrete_function.h
- tf_concrete_function_revival_state.h
- tf_signature_def_function.cc
- tf_signature_def_function.h
- tf_signature_def_function_revival_state.h
- variable.cc
- variable.h
- BUILD
- concrete_function.h
- constant_loading_test.cc
- function_metadata.h
- object_graph_traversal_test.cc
- saved_model_api.h
- saved_model_utils.cc
- saved_model_utils.h
- saved_variable_loading_test.cc
- signature_def_function.h
- signature_def_function_metadata.cc
- signature_def_function_metadata.h
- signature_flattening_test.cc
- tensor_spec.cc
- tensor_spec.h
- test_utils.cc
- test_utils.h
- tf_concrete_function_loading_test.cc
- tf_concrete_function_test_protos.cc
- tf_concrete_function_test_protos.h
- tf_saved_model_api.cc
- tf_saved_model_api.h
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- BUILD
- gen_saved_models.py
- BUILD
- concrete_function.cc
- concrete_function_list.cc
- concrete_function_list_type.h
- concrete_function_type.h
- function_metadata.cc
- function_metadata_type.h
- saved_model_api.cc
- saved_model_api_test.cc
- saved_model_api_type.h
- signature_def_function.cc
- signature_def_function_metadata.cc
- signature_def_function_metadata_type.h
- signature_def_function_type.h
- signature_def_param.cc
- signature_def_param_list.cc
- signature_def_param_list_type.h
- signature_def_param_type.h
- tensor_spec.cc
- tensor_spec_type.h
- BUILD
- c_saved_model_api.h
- concrete_function.h
- concrete_function_list.h
- function_metadata.h
- README.md
- saved_model_api.h
- signature_def_function.h
- signature_def_function_metadata.h
- signature_def_param.h
- signature_def_param_list.h
- tensor_spec.h
- README.md
- BUILD
- test_pluggable_device.cc
- BUILD
- stream_executor.cc
- stream_executor.h
- stream_executor_internal.h
- stream_executor_test.cc
- stream_executor_test_util.cc
- stream_executor_test_util.h
- bitcast.cc
- histogram_summary.cc
- merge_summary.cc
- summary.cc
- bitcast_op.cc
- bitcast_op_test.cc
- BUILD
- histogram_summary_op.cc
- merge_summary_op.cc
- summary_op.cc
- summary_op_benchmark_test.cc
- summary_op_test.cc
- tensor_shape_utils.cc
- tensor_shape_utils.h
- tensor_shape_utils_test.cc
- tf_record
- BUILD
- c_api.cc
- c_api.h
- c_api_experimental.cc
- c_api_experimental.h
- c_api_experimental_test.cc
- c_api_function.cc
- c_api_function_test.cc
- c_api_internal.h
- c_api_macros.h
- c_api_macros_internal.h
- c_api_test.cc
- c_op_requires.h
- c_test.c
- c_test_util.cc
- c_test_util.h
- checkpoint_reader.cc
- checkpoint_reader.h
- conversion_macros.h
- env.cc
- env.h
- env_test.cc
- exported_symbols.lds
- generate-pc.sh
- kernels.cc
- kernels.h
- kernels_experimental.cc
- kernels_experimental.h
- kernels_test.cc
- ops.cc
- ops.h
- ops_test.cc
- python_api.cc
- python_api.h
- README.md
- safe_ptr.cc
- safe_ptr.h
- tensor_interface.h
- test_op.cc
- test_op1.cc
- tf_attrtype.h
- tf_buffer.cc
- tf_buffer.h
- tf_buffer_internal.h
- tf_datatype.cc
- tf_datatype.h
- tf_file_statistics.h
- tf_shape.cc
- tf_shape.h
- tf_shape_internal.h
- tf_status.cc
- tf_status.h
- tf_status_helper.cc
- tf_status_helper.h
- tf_status_helper_test.cc
- tf_status_internal.h
- tf_tensor.cc
- tf_tensor.h
- tf_tensor_helper.h
- tf_tensor_internal.h
- tf_tstring.cc
- tf_tstring.h
- version_script.lds
- while_loop_test.cc
- client_session.cc
- client_session.h
- client_session_test.cc
- BUILD
- runtime.h
- runtime_builder.h
- status.h
- tensor.h
- tensorhandle.h
- BUILD
- tensor_test.cc
- tensor_types_test_util.h
- tensorhandle_test.cc
- BUILD
- load.cc
- load.h
- load_test.cc
- save.cc
- save.h
- save_test.cc
- BUILD
- cc_op_fuzz_gen.cc
- cc_op_fuzz_gen.h
- cc_op_fuzz_gen_main.cc
- op_fuzzing.bzl
- cc_op_gen.cc
- cc_op_gen.h
- cc_op_gen_main.cc
- cc_op_gen_test.cc
- cc_op_gen_util.cc
- cc_op_gen_util.h
- cc_ops_test.cc
- grad_op_registry.cc
- grad_op_registry.h
- gradient_checker.cc
- gradient_checker.h
- gradient_checker_test.cc
- gradients.cc
- gradients.h
- gradients_test.cc
- ops.cc
- ops.h
- scope.cc
- scope.h
- scope_internal.h
- scope_test.cc
- test_op.cc
- testutil.cc
- testutil.h
- while_gradients.cc
- while_gradients.h
- while_gradients_test.cc
- array_grad.cc
- array_grad_test.cc
- data_flow_grad.cc
- data_flow_grad_test.cc
- functional_grad.cc
- functional_grad_test.cc
- grad_helper.cc
- grad_helper.h
- grad_testutil.cc
- grad_testutil.h
- image_grad.cc
- image_grad_test.cc
- linalg_grad.cc
- linalg_grad_test.cc
- manip_grad.cc
- manip_grad_test.cc
- math_grad.cc
- math_grad_test.cc
- nn_grad.cc
- nn_grad_test.cc
- README.md
- resource_variable_grad.cc
- resource_variable_grad_test.cc
- const_op.cc
- const_op.h
- const_op_test.cc
- standard_ops.h
- while_loop.cc
- while_loop.h
- while_loop_test.cc
- BUILD
- concrete_function.h
- concrete_function_list.h
- function_metadata.h
- saved_model_api.h
- signature_def_function.h
- signature_def_function_metadata.h
- BUILD
- saved_model_api_test.cc
- BUILD
- internal_api.cc
- internal_api.h
- README.md
- BUILD
- loader.clif
- test_asset.txt
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- saved_model.pb
- saved_model.pb
- saved_model.cpb
- saved_model.pbtxt
- variables.data-00000-of-00001
- variables.index
- fingerprint.pb
- saved_model.pb
- saved_model_debug_info.pb
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- empty
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- empty
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- empty
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- foo.txt
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- foo.txt
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- foo.txt
- variables.data-00000-of-00001
- variables.index
- saved_model.pbtxt
- foo.txt
- validation_result_do_not_edit
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- static_hashtable_asset.txt
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- saved_model_debug_info.pb
- variables.data-00000-of-00001
- variables.index
- fingerprint.pb
- saved_model.pb
- saved_model_debug_info.pb
- variables.data-00000-of-00001
- variables.index
- saved_model.pb
- generate_chunked_models.py
- generate_saved_models.py
- static_hashtable_asset.txt
- test_asset.txt
- BUILD
- bundle_v2.cc
- bundle_v2.h
- bundle_v2_test.cc
- constants.h
- fingerprinting.cc
- fingerprinting.h
- fingerprinting_chunked_test.cc
- fingerprinting_test.cc
- fingerprinting_utils.cc
- fingerprinting_utils.h
- fingerprinting_utils_test.cc
- fingerprinting_x_platform_utils.cc
- fingerprinting_x_platform_utils.h
- loader.cc
- loader.h
- loader_util.cc
- loader_util.h
- metrics.cc
- metrics.h
- metrics_test.cc
- reader.cc
- reader.h
- reader_test.cc
- README.md
- saved_model_bundle_lite_test.cc
- saved_model_bundle_test.cc
- saved_model_fuzz.cc
- signature_constants.h
- singleprint.cc
- singleprint.h
- singleprint_test.cc
- tag_constants.h
- test_utils.h
- util.cc
- util.h
- util_test.cc
- BUILD
- freeze_saved_model.cc
- freeze_saved_model.h
- freeze_saved_model_test.cc
- coordinator.cc
- coordinator.h
- coordinator_test.cc
- queue_runner.cc
- queue_runner.h
- queue_runner_test.cc
- BUILD
- bench_graph_tfmatmul.template.pbtxt
- BUILD
- make_test_graphs.py
- test_error_message.lit.pbtxt
- test_error_message.lit.pbtxt.config.pbtxt
- test_error_message.lit.pbtxt.debug.pbtxt
- test_error_message.lit.pbtxt.fake_py.debug
- test_graph_tfadd.config.pbtxt
- test_graph_tfadd_with_ckpt.config.pbtxt
- test_graph_tfassert_eq.config.pbtxt
- test_graph_tfcond.config.pbtxt
- test_graph_tffunction.config.pbtxt
- test_graph_tfgather.config.pbtxt
- test_graph_tfmatmul.config.pbtxt
- test_graph_tfmatmul_with_constant.config.pbtxt
- test_graph_tfmatmulandadd.config.pbtxt
- test_graph_tfrandom_uniform.config.pbtxt
- test_graph_tfscatter.config.pbtxt
- test_graph_tfsplits.config.pbtxt
- test_graph_tftop_k.config.pbtxt
- test_graph_tfvariable.config.pbtxt
- test_graph_tfvariable_readonly.config.pbtxt
- test_graph_tfvariable_sequential_updates.config.pbtxt
- tfcompile_test.cc
- aot_only_var_handle_op.cc
- aot_only_var_handle_op.h
- benchmark.cc
- benchmark.h
- benchmark_main.template
- benchmark_test.cc
- BUILD
- codegen.cc
- codegen.h
- codegen_test.cc
- compile.cc
- compile.h
- embedded_protocol_buffers.cc
- embedded_protocol_buffers.h
- flags.cc
- flags.h
- quantize.h
- test.cc
- test_google.cc
- test_graph_tfadd.config.pbtxt
- test_graph_tfadd.pbtxt
- test_graph_tfunknownop.config.pbtxt
- test_graph_tfunknownop.pbtxt
- test_graph_tfunknownop2.config.pbtxt
- test_graph_tfunknownop3.config.pbtxt
- tfcompile.bzl
- tfcompile_main.cc
- thunk_proto_execution_deserializer.cc
- thunk_proto_execution_deserializer.h
- BUILD
- xla_ops.cc
- xla_ops.h
- BUILD
- xla_ops.cc
- xla_ops_grad.py
- auto_clustering_test.cc
- auto_clustering_test_helper.cc
- auto_clustering_test_helper.h
- BUILD
- device_compiler_serialize_options_test.cc
- device_compiler_serialize_test.cc
- device_compiler_test_helper.cc
- device_compiler_test_helper.h
- keras_imagenet_main.golden_summary
- keras_imagenet_main.pbtxt
- keras_imagenet_main_graph_mode.golden_summary
- keras_imagenet_main_graph_mode.pbtxt
- opens2s_gnmt_mixed_precision.golden_summary
- opens2s_gnmt_mixed_precision.pbtxt.gz
- BUILD
- build_xla_ops_pass.cc
- build_xla_ops_pass.h
- build_xla_ops_pass_test.cc
- clone_constants_for_better_clustering.cc
- clone_constants_for_better_clustering.h
- clone_constants_for_better_clustering_test.cc
- cluster_scoping_pass.cc
- cluster_scoping_pass.h
- cluster_scoping_pass_test.cc
- compilability_check_util.cc
- compilability_check_util.h
- compilability_check_util_test.cc
- compilation_passes_test_main.cc
- deadness_analysis.cc
- deadness_analysis.h
- deadness_analysis_internal.h
- deadness_analysis_test.cc
- defs.cc
- defs.h
- device_compilation_cache.h
- device_compilation_cache_test.cc
- device_compilation_cluster_signature.cc
- device_compilation_cluster_signature.h
- device_compilation_cluster_signature_test.cc
- device_compilation_profiler.cc
- device_compilation_profiler.h
- device_compilation_profiler_test.cc
- device_compiler.h
- device_compiler_client.cc
- device_compiler_client.h
- device_compiler_client_test.cc
- device_compiler_disable_test.cc
- device_compiler_test.cc
- device_context_test.cc
- device_executable_persistor.cc
- device_executable_persistor.h
- device_executable_persistor_test.cc
- device_util.cc
- device_util.h
- device_util_test.cc
- encapsulate_subgraphs_pass.cc
- encapsulate_subgraphs_pass.h
- encapsulate_subgraphs_pass_test.cc
- encapsulate_util.cc
- encapsulate_util.h
- encapsulate_util_test.cc
- encapsulate_xla_computations_pass.cc
- encapsulate_xla_computations_pass.h
- encapsulate_xla_computations_pass_test.cc
- extract_outside_compilation_pass.cc
- extract_outside_compilation_pass.h
- extract_outside_compilation_pass_test.cc
- flags.cc
- flags.h
- force_xla_constants_on_host_pass.cc
- force_xla_constants_on_host_pass.h
- force_xla_constants_on_host_pass_test.cc
- get_compiler_ir.cc
- get_compiler_ir.h
- increase_dynamism_for_auto_jit_pass.cc
- increase_dynamism_for_auto_jit_pass.h
- increase_dynamism_for_auto_jit_pass_test.cc
- jit_compilation_pass_registration.cc
- mark_for_compilation_pass.cc
- mark_for_compilation_pass.h
- mark_for_compilation_pass_test.cc
- mark_for_compilation_pass_test_helper.cc
- mark_for_compilation_pass_test_helper.h
- node_matchers.cc
- node_matchers.h
- node_matchers_test.cc
- package_groups.bzl
- partially_decluster_pass.cc
- partially_decluster_pass.h
- partially_decluster_pass_test.cc
- pjrt_base_device.cc
- pjrt_base_device.h
- pjrt_compile_util.cc
- pjrt_compile_util.h
- pjrt_compile_util_test.cc
- pjrt_device_compiler_client.cc
- pjrt_device_compiler_client.h
- pjrt_device_context.cc
- pjrt_device_context.h
- pjrt_tensor_buffer.h
- pjrt_tensor_buffer_util.cc
- pjrt_tensor_buffer_util.h
- pjrt_tensor_buffer_util_test.cc
- rearrange_function_argument_pass_test.cc
- report_clustering_info_pass.cc
- report_clustering_info_pass.h
- resource_operation_safety_analysis.cc
- resource_operation_safety_analysis.h
- resource_operation_safety_analysis_test.cc
- shape_inference.cc
- shape_inference.h
- shape_inference_helpers.cc
- shape_inference_helpers.h
- shape_inference_test.cc
- test_util.cc
- test_util.h
- tf_graph_to_hlo_compiler.cc
- tf_graph_to_hlo_compiler.h
- tf_graph_to_hlo_compiler.proto
- tf_to_hlo_compiler.h
- variable_info.cc
- variable_info.h
- variable_info_util.cc
- variable_info_util.h
- xla_activity.proto
- xla_activity_listener.cc
- xla_activity_listener.h
- xla_activity_listener_test.cc
- xla_activity_logging_listener.cc
- xla_cluster_util.cc
- xla_cluster_util.h
- xla_cluster_util_test.cc
- xla_compilation_cache.proto
- xla_compile_on_demand_op.cc
- xla_compile_on_demand_op.h
- xla_compile_util.cc
- xla_compile_util.h
- xla_compile_util_test.cc
- xla_compiler_options_util.cc
- xla_compiler_options_util.h
- xla_compiler_options_util_test.cc
- xla_cpu_device.cc
- xla_device.cc
- xla_device.h
- xla_device_compiler_client.cc
- xla_device_compiler_client.h
- xla_device_context.cc
- xla_device_context.h
- xla_device_ops.cc
- xla_device_ops.h
- xla_gpu_device.cc
- xla_host_recv_device_context.cc
- xla_host_recv_device_context.h
- xla_host_send_device_context.cc
- xla_host_send_device_context.h
- xla_host_send_recv_device_context_test.cc
- xla_kernel_creator.cc
- xla_kernel_creator.h
- xla_kernel_creator_test.cc
- xla_launch_util.cc
- xla_launch_util.h
- xla_launch_util_gpu_test.cc
- xla_launch_util_test.cc
- xla_ops_on_regular_devices.cc
- xla_platform_info.cc
- xla_platform_info.h
- xla_platform_info_test.cc
- xla_tensor.cc
- xla_tensor.h
- xla_tpu_device.cc
- tf_passes.md
- mlir-infra.svg
- _book.yaml
- _index.yaml
- dialects.md
- overview.md
- README.md
- tf_passes.md
- BUILD
- padding.h
- tfl_pass_config.h
- BUILD
- error_reporter.cc
- error_reporter.h
- error_reporter_test.cc
- flatbuffer_conversions.cc
- flatbuffer_conversions.h
- flatbuffer_conversions_test.cc
- verifier.h
- BUILD
- builtin_op_data.h
- tflite_types.h
- absl_error_model_builder.cc
- absl_error_model_builder.h
- BUILD
- macros.h
- model_builder_base.cc
- model_builder_base.h
- special_rules.bzl
- BUILD
- debug.cc
- debug.h
- debug_options.proto
- debug_test.cc
- allowlisted_flex_ops.cc
- allowlisted_flex_ops.h
- allowlisted_flex_ops_internal.h
- BUILD
- build_def.bzl
- BUILD
- BUILD
- outline_operations.cc
- outline_operations.h
- BUILD
- metadata_util.cc
- metadata_util.h
- metadata_util_test.cc
- rematerializer.cc
- rematerializer.h
- rematerializer_test.cc
- cost.h
- subgraph.h
- targets.h
- utils.cc
- utils.h
- BUILD
- example_hardware.cc
- example_hardware.h
- compute_cost.png
- final_graph.png
- pick_subgraphs.png
- target_annotation.png
- target_optimization.png
- BUILD
- cpu_hardware.cc
- gpu_hardware.cc
- gpu_hardware.h
- nnapi_hardware.cc
- nnapi_hardware.h
- simple_hardware.cc
- simple_hardware.h
- target_hardware.cc
- target_hardware.h
- _pywrap_tac_wrapper.pyi
- BUILD
- tac_wrapper.cc
- tac_wrapper.h
- tac_wrapper_pybind11.cc
- BUILD
- device-transform-nnapi.mlir
- simple-graph.mlir
- BUILD
- compute-cost.mlir
- device-transform-gpu.mlir
- device-transform-nnapi.mlir
- fold-constants-to-subgraph.mlir
- get-alternative-subgraph.mlir
- get-op-cost.mlir
- pick-subgraphs.mlir
- raise-target-subgraphs.mlir
- tac-filter.mlir
- target-annotation.mlir
- compute_cost.cc
- cost_model.cc
- cost_model.h
- device_transform.cc
- device_transform.h
- device_transform_gpu.cc
- device_transform_gpu.h
- device_transform_nnapi.cc
- device_transform_patterns.cc
- device_transform_patterns.h
- fold_constants_to_subgraph.cc
- get_alternative_subgraph.cc
- passes.h
- pick_subgraphs.cc
- raise_target_subgraphs.cc
- tac_filter.cc
- tac_pass.h
- target_annotation.cc
- transform_patterns.td
- BUILD
- utils.cc
- utils.h
- BUILD
- execution_metadata_exporter.cc
- execution_metadata_exporter.h
- execution_metadata_exporter_test.cc
- README.md
- runtime_metadata.fbs
- tac.py
- tac_filter.proto
- tac_importer_exporter.h
- tac_module.cc
- tac_module.h
- tac_translate.cc
- tflite_import_export.cc
- tflite_import_export.h
- tfl_canonicalize.td
- tfl_op_enums.td
- tfl_op_interfaces.td
- tfl_ops.cc
- tfl_ops.h
- tfl_ops.td
- neon_check.h
- BUILD
- sparsity_format_converter.cc
- sparsity_format_converter.h
- BUILD
- common.cc
- common.h
- compatibility_macros.h
- cppmath.h
- quantization_util.cc
- quantization_util.h
- README
- runtime_shape.cc
- runtime_shape.h
- runtime_shape_test.cc
- BUILD
- padding.h
- strided_slice.mlir
- BUILD
- converter_error_data.proto
- error_collector.cc
- error_collector.h
- error_collector_inst.cc
- error_collector_inst.h
- error_collector_inst_test.cc
- types_util.cc
- types_util.h
- BUILD
- python_error_reporter.cc
- python_error_reporter.h
- python_utils.cc
- python_utils.h
- _pywrap_converter_api.pyi
- BUILD
- converter_python_api.cc
- converter_python_api.h
- converter_python_api_wrapper.cc
- flatbuffer_to_mlir.cc
- flatbuffer_to_mlir.h
- graphdef_to_tfl_flatbuffer.cc
- graphdef_to_tfl_flatbuffer.h
- jax_to_tfl_flatbuffer.cc
- jax_to_tfl_flatbuffer.h
- saved_model_to_tfl_flatbuffer.cc
- saved_model_to_tfl_flatbuffer.h
- tf_tfl_flatbuffer_helpers.cc
- tf_tfl_flatbuffer_helpers.h
- wrap_converter.py
- BUILD
- quantization.td
- quantization_config.cc
- quantization_config.h
- quantization_driver.cc
- quantization_driver.h
- quantization_driver_test.cc
- quantization_traits.h
- quantization_utils.cc
- quantization_utils.h
- tfl_quantization_driver.cc
- tfl_quantization_driver.h
- BUILD
- test_base.h
- .clang-format
- .clang-tidy
- BUILD
- ConvertConst.cc
- ConvertSimQuant.cc
- Passes.h
- Passes.td
- QuantizeUtils.cc
- QuantizeUtils.h
- QuantOps.cc
- QuantOps.h
- QuantOps.td
- QuantOpsBase.td
- add_with_const_input.bin
- argmax.bin
- broadcast_to.bin
- concat.bin
- custom_op.bin
- fc.bin
- fc_qat.bin
- gather_nd.bin
- lstm_calibrated.bin
- lstm_calibrated2.bin
- lstm_quantized.bin
- lstm_quantized2.bin
- maximum.bin
- minimum.bin
- mixed.bin
- mixed16x8.bin
- multi_input_add_reshape.bin
- pack.bin
- quantized_with_gather.bin
- README.md
- resource_vars_calibrated.bin
- single_avg_pool_min_minus_5_max_plus_5.bin
- single_conv_no_bias.bin
- single_conv_weights_min_0_max_plus_10.bin
- single_conv_weights_min_minus_127_max_plus_127.bin
- single_softmax_min_minus_5_max_plus_5.bin
- split.bin
- svdf_calibrated.bin
- svdf_quantized.bin
- transpose.bin
- unidirectional_sequence_lstm_calibrated.bin
- unidirectional_sequence_lstm_quantized.bin
- unpack.bin
- weight_shared_between_convs.bin
- where.bin
- BUILD
- model_utils.cc
- model_utils.h
- model_utils_test.cc
- portable_tensor_utils.cc
- portable_tensor_utils.h
- quantization_utils.cc
- quantization_utils.h
- quantization_utils_test.cc
- quantize_weights.cc
- quantize_weights.h
- quantize_weights_portable.cc
- quantize_weights_test.cc
- README
- BUILD
- quantize_model.cc
- quantize_model.h
- quantize_model_test.cc
- quantize_weights.cc
- quantize_weights.h
- quantize_weights_test.cc
- test_util.cc
- test_util.h
- tfl_quantizer.cc
- tfl_to_std.cc
- tfl_to_std.h
- BUILD
- fallback_to_flex_ops_default.mlir
- fallback_to_flex_ops_legacy.mlir
- tf_to_quant.mlir
- tf_to_quant_4bit.mlir
- BUILD
- fallback_to_flex_ops.cc
- fallback_to_flex_patterns.td
- passes.h
- tf_to_quant.cc
- BUILD
- import_quant_stats.mlir
- op_quant_spec_getters_gen.cc
- tflite_op_coverage_spec_getters_gen.cc
- BUILD
- device_target.cc
- device_target.h
- import_quant_stats_pass.cc
- numerical_utils.cc
- numerical_utils.h
- numerical_utils_test.cc
- quantization_context.cc
- quantization_context.h
- quantization_info.proto
- quantization_passes.h
- BUILD
- conversion_metadata.fbs
- conversion_metadata_generated.h
- debug_metadata.fbs
- flatbuffer_compatibility_test.cc
- README.md
- schema.fbs
- schema_conversion_utils.cc
- schema_conversion_utils.h
- schema_generated.h
- schema_utils.cc
- schema_utils.h
- schema_v0.fbs
- schema_v1.fbs
- schema_v2.fbs
- schema_v3.fbs
- schema_v3a.fbs
- schema_v3b.fbs
- schema_v3c.fbs
- sparse_tensor.bin
- BUILD
- sparsify_model.cc
- sparsify_model.h
- sparsify_model_test.cc
- BUILD
- shlo_simplify.mlir
- shlo_simplify.cc
- shlo_simplify.td
- BUILD
- folders.cc
- folders.h
- odml_converter_main.cc
- passes.h
- passes.td
- BUILD
- call_xla_module_to_stablehlo.mlir
- compose-uniform-quantized-type.mlir
- composite-lowering.mlir
- drop_shape_assertions.mlir
- fold_broadcast.mlir
- fuse_mhlo_convolution.mlir
- gelu.mlir
- legalize-inplaceupdate.mlir
- legalize-skip-partitioned-calls.mlir
- legalize-skip-quantization-ops.mlir
- legalize-stablehlo-tfl-composite-quantize-dequantize.mlir
- legalize-stablehlo-tfl-composite.mlir
- legalize-stablehlo-vhlo.mlir
- legalize-tf.mlir
- legalize-tfl-stablehlo-add.mlir
- legalize-tfl-stablehlo-broadcast.mlir
- legalize-tfl-stablehlo-clamp.mlir
- legalize-tfl-stablehlo-concat.mlir
- legalize-tfl-stablehlo-constant.mlir
- legalize-tfl-stablehlo-conv.mlir
- legalize-tfl-stablehlo-max.mlir
- legalize-tfl-stablehlo-mul.mlir
- legalize-tfl-stablehlo-pad.mlir
- legalize-tfl-stablehlo-reshape.mlir
- legalize-tfl-stablehlo-rsqrt.mlir
- legalize-tfl-stablehlo-sub.mlir
- legalize-tfl-stablehlo.mlir
- legalize_hlo.mlir
- odml-to-stablehlo-allow-tf.mlir
- odml-to-stablehlo-smuggle-resize.mlir
- optimize.mlir
- optimize_layout.mlir
- prepare_hlo.mlir
- stablehlo-custom-call-legalize-composite.mlir
- stablehlo_fuse_convolution.mlir
- stablehlo_unfuse_batch_norm.mlir
- tf-tfl-translate-serialize-stablehlo-clamp.mlir
- tf-tfl-translate-serialize-stablehlo-concat.mlir
- tf-tfl-translate-serialize-stablehlo-conv.mlir
- tf-tfl-translate-serialize-stablehlo-division.mlir
- tf-tfl-translate-serialize-stablehlo-logistic.mlir
- tf-tfl-translate-serialize-stablehlo-multiply.mlir
- tf-tfl-translate-serialize-stablehlo-resize-bilinear.mlir
- tf-tfl-translate-serialize-stablehlo.mlir
- tf-tfl-translate-tf-quantize.mlir
- tfl_legalize_chlo.mlir
- tfl_legalize_hlo.mlir
- tfl_legalize_hlo_custom_call.mlir
- torch-build-stablehlo-composite.mlir
- unfold_splat_constant_pass.mlir
- unfuse_mhlo_batch_norm.mlir
- uniform-quantized-stablehlo-to-tfl.mlir
- BUILD
- case.cc
- case.h
- conv.cc
- conv.h
- conv_util.cc
- conv_util.h
- custom_call.cc
- custom_call.h
- dot_general.cc
- dot_general.h
- fft.cc
- fft.h
- gather.cc
- gather.h
- gelu.cc
- gelu.h
- get_dimension_size.cc
- get_dimension_size.h
- if.cc
- if.h
- iota.cc
- iota.h
- op_util_common.cc
- op_util_common.h
- pad.cc
- pad.h
- pad_util.cc
- pad_util.h
- reduce.cc
- reduce.h
- reduce_window.cc
- reduce_window.h
- reduce_window_util.cc
- reduce_window_util.h
- scatter.cc
- scatter.h
- slice.cc
- slice.h
- sort.cc
- sort.h
- util.cc
- util.h
- while.cc
- while.h
- README.md
- build_stablehlo_composite_pass.cc
- legalize_qdq_custom_call_pass.cc
- lift_callsite_loc_caller_pass.cc
- check_accepted_ops_pass.cc
- check_accepted_ops_pass.h
- compose_uniform_quantized_type_pass.cc
- composite_avg_pool.cc
- composite_avg_pool.h
- composite_avg_pool_patterns.td
- composite_lowering_pass.cc
- composite_lowering_pass.h
- composite_lowering_patterns.td
- composite_utils.cc
- composite_utils.h
- composite_utils.td
- drop_savedmodel_semantics.cc
- drop_savedmodel_semantics.h
- drop_shape_assertions.cc
- hlo_matchers.cc
- hlo_matchers.h
- legalize_hlo.cc
- legalize_hlo_patterns.td
- legalize_stablehlo_composite_to_tfl_custom.cc
- legalize_stablehlo_custom_call_to_composite.cc
- legalize_stablehlo_to_vhlo.cc
- legalize_tf_xla_call_module_to_stablehlo_pass.cc
- legalize_tf_xla_call_module_to_stablehlo_pass.h
- op_stat_pass.cc
- op_stat_pass.h
- optimize.cc
- optimize_layout.cc
- prepare_hlo.cc
- prepare_hlo.td
- smuggle_disallowed_ops.cc
- smuggle_disallowed_ops.h
- stablehlo_fuse_convolution_pass.cc
- stablehlo_passes.h
- stablehlo_passes.td
- stablehlo_unfuse_batch_norm_pass.cc
- stablehlo_util.cc
- stablehlo_util.h
- tfl_legalize_chlo.cc
- tfl_stablehlo_pass.cc
- tfl_stablehlo_pass.h
- tflite_legalize_hlo.cc
- tflite_legalize_hlo_patterns.td
- transforms.cc
- transforms.h
- unfold_splat_constant_pass.cc
- uniform_quantized_stablehlo_to_tfl_pass.cc
- BUILD
- odml_to_stablehlo.cc
- add.bin
- multi_signatures.bin
- BUILD
- concrete_function_error.py
- saved_model_error.py
- v1_1.0_224_frozen.wrong_attr.line.part.pbtxt
- v1_1.0_224_frozen.wrong_attr.line.part.pbtxt.debug
- v1_1.0_224_frozen.wrong_attr.stack.part.pbtxt
- v1_1.0_224_frozen.wrong_attr.stack.part.pbtxt.debug
- add.pbtxt
- BUILD
- control_flow_v1.pbtxt
- custom_opdef.pbtxt
- disallow_stateful_partitioned_call.pbtxt
- fake_quant_per_channel.pbtxt
- fake_quant_per_channel_4bit.pbtxt
- fake_quant_without_identity.pbtxt
- fake_quant_without_identity_4bit.pbtxt
- graph-input-node.pbtxt
- graph_with_placeholder_with_default.pbtxt
- if_op.pbtxt
- quant_stats.pbtxt
- quant_stats.pbtxt.stats
- unroll_batch_matmul.pbtxt
- unroll_batch_matmul_disabled.pbtxt
- basic_lstm.mlir
- batched_gather_round_trip.mlir
- batched_scatter_round_trip.mlir
- bucketize.mlir
- BUILD
- cast_bf16.mlir
- composite_op_round_trip.mlir
- constants.mlir
- constants_offset.mlir
- control_edges.mlir
- custom_op.mlir
- custom_op_offset.mlir
- debug_metadata.mlir
- dense_constants.mlir
- dense_constants_offset.mlir
- dynamic_shape.mlir
- empty_input_output_names.json
- external_buffer.mlir
- external_constant.mlir
- if_op.mlir
- import_json.json
- importer_test_min_max.cc
- importer_test_min_max.cc.mlir
- input_arrays.mlir
- input_output_names_attr.mlir
- legacy_reshape.json
- lstm.json
- lstm.mlir
- many_attribute_op.mlir
- math.mlir
- matmul.mlir
- mix_tflite_vhlo.mlir
- multi_output_op.json
- op_location_with_branches.mlir
- optional.mlir
- optional_input.json
- output_arrays.mlir
- pruning.mlir
- pruning_function_input_as_output.mlir
- quant_stats.mlir
- quantization.mlir
- reshape.mlir
- signature.mlir
- signature_with_multiple_entry_points.mlir
- simple.mlir
- test_schema.fbs
- tf_variant_type.mlir
- unranked_function_output.mlir
- unranked_tensor.mlir
- variable.mlir
- vhlo.mlir
- vhlo_const.mlir
- vhlo_custom_call.mlir
- while_op.mlir
- basic_lstm.mlir
- bucketize.mlir
- BUILD
- cast_bf16.mlir
- composite.mlir
- custom_op_with_tflite_op.mlir
- custom_tensorlist_reserve.mlir
- debug_metadata.mlir
- deduplicate_const.mlir
- depthwise_conv2d.mlir
- depthwise_conv2d_v2.mlir
- disable_builtin.mlir
- disable_custom.mlir
- disable_flex.mlir
- disable_flex_enable_builtin.mlir
- dynamic_shape_constant.mlir
- external_buffer.mlir
- fake_quant.mlir
- flex_exclusively.mlir
- flex_op_with_complex128.mlir
- flex_op_with_f64.mlir
- flex_op_with_tflite_op.mlir
- fully_connected.mlir
- fully_connected_v2.mlir
- hashtable_resource.mlir
- if_op.mlir
- logical.mlir
- low_bit_packing.mlir
- lstm.mlir
- lstm_asym_attr.mlir
- lstm_quantized.mlir
- math.mlir
- metadata.mlir
- mul_v2.mlir
- mul_v3.mlir
- nn.mlir
- numeric_verify.mlir
- optional.mlir
- quantization.mlir
- reshape.mlir
- runtime_version_metadata.mlir
- signature_def.mlir
- signature_def_output_override.mlir
- signature_def_with_multiple_entry_points.mlir
- signature_def_with_no_inputs.mlir
- simple.mlir
- simple_with_connected_control_nodes.mlir
- simple_with_unconnected_control_nodes.mlir
- svdf.mlir
- svdf_v2.mlir
- tf_entry_function.mlir
- tfl_while_op.mlir
- transpose_conv_optional.mlir
- type_attr.mlir
- u16_quant.mlir
- unidirectional_sequence_lstm.mlir
- unidirectional_sequence_rnn.mlir
- unranked_tensor.mlir
- unsorted_segment_prod.mlir
- variable.mlir
- variant_type_on_func.mlir
- variant_type_on_op.mlir
- while_op.mlir
- analyze-variables.mlir
- bias-quantizer.mlir
- BUILD
- canonicalize.mlir
- canonicalize_boundary_value.mlir
- cast_bf16_ops_to_f32.mlir
- cleanup_optimization_barrier.mlir
- const-fold-binary-ops.mlir
- const-fold-unary-resource.mlir
- const-fold.mlir
- decompose-hybrid-quantization.mlir
- default_quant_params.mlir
- dilated-conv.mlir
- downcast-x64.mlir
- elements-attr-roundtrip.mlir
- fuse-qdq.mlir
- fuse-tftext.mlir
- get-arithmetic-count.mlir
- gpu_compatibility.mlir
- guarantee_func_has_one_use.mlir
- inlining.mlir
- insert_call_once_op.mlir
- large_constant_fold.mlir
- legalize-tensorlist.mlir
- legalize-tf-assert.mlir
- legalize-tf-hashtables.mlir
- legalize-tf-no-runtime-verification.mlir
- legalize-tf-variables.mlir
- legalize-tf-while.mlir
- legalize-tf.mlir
- legalize_jax_random.mlir
- lift_tflite_flex_ops.mlir
- lower-static-tensor-list-default-to-single-batch.mlir
- lower-static-tensor-list-enable-dynamic-update-slice.mlir
- lower-static-tensor-list.mlir
- lower_quant_annotations.mlir
- modify_io_nodes.mlir
- ops.mlir
- optimize-after-quantization.mlir
- optimize.mlir
- optimize_batch_matmul.mlir
- optimize_broadcast_like.mlir
- optimize_functional_ops.mlir
- optimize_gather_nd_to_slice.mlir
- optimize_no_verify.mlir
- optimize_op_order.mlir
- partitioned-topological-sort.mlir
- pin-ops-with-side-effects.mlir
- post-quantize-dynamic-range.mlir
- post-quantize.mlir
- prepare-composite-functions-tf.mlir
- prepare-quantize-dynamic-range.mlir
- prepare-quantize-post-training-16bits.mlir
- prepare-quantize-post-training.mlir
- prepare-quantize-signed.mlir
- prepare-quantize.mlir
- prepare-tf-fake-quant-4bit.mlir
- prepare-tf-fake-quant.mlir
- prepare-tf-with-allowing-bf16-and-f16-type-legalization.mlir
- prepare-tf.mlir
- propagate-qsv.mlir
- push-tpose-through-ewise.mlir
- quantize-dynamic-range-float16.mlir
- quantize-dynamic-range.mlir
- quantize-numeric-verify.mlir
- quantize-strict.mlir
- quantize-variables.mlir
- quantize.mlir
- raise-custom-ops.mlir
- rank_reduction.mlir
- reduce-type-precision.mlir
- reduce_while_operands.mlir
- shape-inference.mlir
- split-merged-operands.mlir
- tfl_if_outline.mlir
- tfl_while_op_licm.mlir
- tfl_while_outline.mlir
- trim-functions-tf.mlir
- unfold-large-splat-constant.mlir
- unfreeze_mutable_global_tensors.mlir
- BUILD
- operator_property.cc
- operator_property.h
- reduced_precision_metadata.h
- BUILD
- op_signature.cc
- op_signature.h
- op_signature_test.cc
- op_version.cc
- op_version.h
- op_version_test.cc
- runtime_version.cc
- runtime_version.h
- runtime_version_test.cc
- BUILD
- command_line_flags.cc
- command_line_flags.h
- bias_quantizer_pass.cc
- fuse_qdq_pass.cc
- propagate_qsv_pass.cc
- quant_utils.cc
- quant_utils.h
- strict_quantize_patterns.td
- analyze_variables_pass.cc
- analyze_variables_pass.h
- legalize_tensorlist_pass.cc
- legalize_tensorlist_pass.h
- while_loop_outline_pass.cc
- while_loop_outline_pass.h
- optimize_batch_matmul_utils.cc
- optimize_batch_matmul_utils.h
- optimize_batch_matmul_utils_test.cc
- split_merged_operands_pass.cc
- split_merged_operands_pass.h
- unfold_large_splat_constants_pass.cc
- unfold_large_splat_constants_pass.h
- elements_attr_roundtrip_pass.cc
- elements_attr_roundtrip_pass.h
- canonicalize_boundary_value_pass.cc
- canonicalize_boundary_value_pass.h
- cast_bf16_ops_to_f32_pass.cc
- cast_bf16_ops_to_f32_pass.h
- cleanup_optimization_barrier_pass.cc
- cleanup_optimization_barrier_pass.h
- converter_pass_options_setter.cc
- converter_pass_options_setter.h
- decompose_hybrid_quantization.cc
- default_quant_params.cc
- dense_to_sparse_pass.cc
- dense_to_sparse_pass.h
- dilated_conv.cc
- dilated_conv.h
- downcast_x64_pass.cc
- downcast_x64_pass.h
- get_arithmetic_count.cc
- gpu_compatibility.cc
- if_outline.cc
- insert_call_once_op.cc
- large_constant_fold_pass.cc
- large_constant_fold_pass.h
- legalize_hashtables.cc
- legalize_jax_random.cc
- legalize_patterns.td
- legalize_tensorlist.td
- legalize_tf.cc
- legalize_tf_while.cc
- legalize_variables.cc
- legalize_variables.td
- lift_tflite_flex_ops.cc
- lift_tflite_flex_ops.h
- lower_quant_annotations_helper.cc
- lower_quant_annotations_helper.h
- lower_quant_annotations_pass.cc
- lower_static_tensor_list.cc
- modify_io_nodes.cc
- optimize_batch_matmul.td
- optimize_batch_matmul_pass.cc
- optimize_batch_matmul_pass.h
- optimize_broadcast_like_pass.cc
- optimize_broadcast_like_pass.h
- optimize_broadcast_like_pass_options.h
- optimize_broadcast_like_patterns.td
- optimize_functional_ops.cc
- optimize_op_order.cc
- optimize_pass.cc
- optimize_pass.h
- optimize_pass_options.h
- optimize_patterns.td
- partitioned_topological_sort.cc
- pass.h
- pass_options.h
- pass_options_setter.h
- pass_registry_utils.h
- passes.h
- passes.td
- pin_ops_with_side_effects.cc
- pipeline.h
- post_quantize.cc
- post_quantize_patterns.td
- prepare_composite_functions_tf.cc
- prepare_patterns.td
- prepare_quantize.cc
- prepare_quantize_dynamic_range.cc
- prepare_quantize_helper.cc
- prepare_quantize_helper.h
- prepare_tf.cc
- push_transpose_through_ewise_pass.cc
- push_transpose_through_ewise_pass.h
- quantize.cc
- quantize_by_converter_patterns.td
- quantize_patterns.td
- quantize_variables.cc
- raise_custom_ops.cc
- rank_reduction.cc
- reduce_type_precision.cc
- reduce_while_operands.cc
- runtime_verify.cc
- tensorlist_patterns.td
- trim_functions_tf.cc
- unfreeze_global_constants.cc
- unfreeze_global_constants.h
- variable_freezing_pipeline.cc
- variable_freezing_pipeline.h
- variable_freezing_pipeline_options.h
- arithmetic_count_util.h
- attribute_utils.cc
- attribute_utils.h
- const_tensor_utils.cc
- const_tensor_utils.h
- constant_utils.cc
- constant_utils.h
- control_edges.h
- convert_type.cc
- convert_type.h
- fake_quant_utils.cc
- fake_quant_utils.h
- low_bit_utils.cc
- low_bit_utils.h
- low_bit_utils_test.cc
- lstm_utils.cc
- lstm_utils.h
- lstm_utils_test.cc
- metadata_utils.h
- mlir_module_utils.h
- nms_utils.cc
- nms_utils.h
- perception_ops_utils.cc
- perception_ops_utils.h
- perception_ops_utils_test.cc
- region_isolation.cc
- region_isolation.h
- region_isolation_test.cc
- shape_and_size_utils.cc
- shape_and_size_utils.h
- shape_and_size_utils_test.cc
- stateful_ops_utils.cc
- stateful_ops_utils.h
- string_utils.cc
- string_utils.h
- tftext_utils.cc
- tftext_utils.h
- tftext_utils_test.cc
- utils.h
- utils.td
- utils_test.cc
- validators.cc
- validators.h
- variables_utils.cc
- variables_utils.h
- allocation.cc
- allocation.h
- BUILD
- build_def.bzl
- converter_flags.proto
- converter_gen.cc
- flatbuffer_export.cc
- flatbuffer_export.h
- flatbuffer_export_flags.h
- flatbuffer_import.cc
- flatbuffer_import.h
- flatbuffer_operator.cc
- flatbuffer_operator.h
- flatbuffer_to_string.cc
- flatbuffer_translate.cc
- flatbuffer_translate.h
- flatbuffer_translate_flags.h
- glob_lit_test.bzl
- json_to_flatbuffer.cc
- litert_mlir_opt_main.cc
- mmap_allocation.cc
- mmap_allocation_disabled.cc
- model_flags.proto
- offset_buffer.h
- offset_buffer_test.cc
- README.md
- register_lite_dialects.cc
- register_lite_dialects.h
- register_lite_dialects_test.cc
- runlit.cfg.py
- runlit.site.cfg.py
- special_rules.bzl
- stateful_error_reporter.h
- symlink_files.bzl
- tf_tfl_passes.cc
- tf_tfl_passes.h
- tf_tfl_translate.cc
- tf_tfl_translate_cl.cc
- tf_tfl_translate_cl.h
- tf_to_tfl_flatbuffer.cc
- tf_to_tfl_flatbuffer.h
- types.proto
- version.h
- attrs.cc
- basic_classes.cc
- BUILD
- builders.cc
- filecheck_wrapper.cc
- filecheck_wrapper.pyi
- mlir_wrapper.cc
- mlir_wrapper.h
- mlir_wrapper.pyi
- ops.cc
- types.cc
- BUILD
- mlir.cc
- mlir.h
- mlir_test.cc
- BUILD
- ConvertConst.cc
- ConvertSimQuant.cc
- FakeQuantSupport.cc
- FakeQuantSupport.h
- Passes.h
- Passes.td
- QuantizeUtils.cc
- QuantizeUtils.h
- QuantOps.cc
- QuantOps.h
- QuantOps.td
- QuantOpsBase.td
- UniformSupport.cc
- UniformSupport.h
- BUILD
- testing.py
- testing_test.py
- BUILD
- portable_tensor_utils.cc
- portable_tensor_utils.h
- quantization.td
- quantization_config.cc
- quantization_config.h
- quantization_driver.cc
- quantization_driver.h
- quantization_driver_test.cc
- quantization_traits.h
- quantization_utils.cc
- quantization_utils.h
- attrs_and_constraints.cc
- attrs_and_constraints.h
- attrs_and_constraints.td
- attrs_and_constraints_test.cc
- BUILD
- func.cc
- func.h
- func_test.cc
- lift_as_function_call.cc
- lift_as_function_call.h
- lift_as_function_call.td
- lift_as_function_call_test.cc
- test_base.h
- tf_uniform_quantized_types.cc
- tf_uniform_quantized_types.h
- uniform_quantized_types.cc
- uniform_quantized_types.h
- uniform_quantized_types_test.cc
- BUILD
- calibration_parameters.h
- calibration_parameters_test.cc
- component.cc
- component.h
- min_max_value.h
- representative_dataset.cc
- representative_dataset.h
- representative_dataset_test.cc
- statistics.cc
- statistics.h
- BUILD
- component.h
- config.cc
- config.h
- config_test.cc
- context.h
- debugger.cc
- debugger.h
- graph_def.h
- graph_def_test.cc
- io.cc
- io.h
- io_test.cc
- pass_pipeline.cc
- pass_pipeline.h
- permutation.h
- permutation_test.cc
- post_calibration.cc
- post_calibration.h
- pre_calibration.cc
- pre_calibration.h
- pre_calibration_test.cc
- report.cc
- report.h
- report_test.cc
- saved_model_export.cc
- saved_model_export.h
- saved_model_export_test.cc
- saved_model_import.cc
- saved_model_import.h
- saved_model_import_test.cc
- static_range_ptq.cc
- static_range_ptq.h
- types.h
- weight_only_ptq.cc
- weight_only_ptq.h
- BUILD
- save_report.cc
- save_report.h
- save_report_test.cc
- BUILD
- stablehlo_op_quant_spec.cc
- stablehlo_op_quant_spec.h
- stablehlo_op_quant_spec_test.cc
- convert_tf_quant_ops_to_mhlo.cc
- convert_tf_quant_to_mhlo_int_test.cc
- convert_tf_quant_types.cc
- convert_tf_quant_types_test.cc
- legalize_tf_quant_test.cc
- optimize.cc
- optimize.td
- passes.cc
- passes.h
- passes.td
- verify_quant_legalization.cc
- passes.h
- passes.td
- test_lift_quantizable_spots_as_functions_with_quantization_specs.cc
- test_post_calibration_component.cc
- test_pre_calibration_component.cc
- test_tf_to_stablehlo_pass.cc
- convert_func_to_bfloat16.cc
- convert_shape_constraint_to_assert.cc
- convert_xla_call_module_op_to_bfloat16.cc
- defer_activation_transpose.cc
- fold_constant_transpose.cc
- insert_calibration_statistics_saver.cc
- insert_weight_param.cc
- lift_quantizable_spots_as_functions.cc
- lift_quantizable_spots_as_functions_fusion.td
- lift_quantizable_spots_as_functions_simple.td
- merge_fusion_with_dequantize.cc
- nchw_convolution_to_nhwc.cc
- optimize_graph.cc
- optimize_graph.td
- passes.h
- passes.td
- post_quantize.cc
- prepare_quantize.cc
- quantization_patterns.cc
- quantization_patterns.h
- quantize.cc
- quantize_composite_functions.cc
- quantize_weight.cc
- remove_sharding_custom_call.cc
- remove_sharding_custom_call.td
- replace_stablehlo_ops_in_main_function_with_xla_call_module_ops.cc
- restore_function_name.cc
- unfuse_mhlo_batch_norm.cc
- unwrap_xla_call_module_op.cc
- xla_call_module_to_call.cc
- quantize_model_test.py
- quantize_model_test_base.py
- stablehlo_quantizer_odml_oss.ipynb
- BUILD
- pywrap_quantization.cc
- pywrap_quantization.pyi
- pywrap_quantization_lib.cc
- pywrap_quantization_lib.h
- quantization.py
- BUILD
- convert-tf-quant-types.mlir
- convert_tf_quant_ops_to_mhlo.mlir
- optimize.mlir
- verify-quant-legalization.mlir
- post_calibration_component.mlir
- pre_calibration_component.mlir
- tf_to_stablehlo.mlir
- prepare_quantize.mlir
- prepare_quantize_int4.mlir
- prepare_quantize_per_channel.mlir
- quantize.mlir
- quantize_op_with_region.mlir
- quantize_same_scale.mlir
- quantize_weight_only.mlir
- convert_func_to_bfloat16.mlir
- convert_xla_call_module_op_to_bfloat16.mlir
- defer_activation_transpose.mlir
- fold_constant_transpose.mlir
- insert_calibration_statistics_saver.mlir
- insert_calibration_statistics_saver_with_skipping.mlir
- insert_weight_param.mlir
- lift_quantizable_spots_as_functions.mlir
- lift_quantizable_spots_as_functions_with_quantization_specs.mlir
- merge-fusion-with-dequantize.mlir
- nchw_convolution_to_nhwc.mlir
- optimize_graph.mlir
- post_quantize.mlir
- quantize_composite_functions.mlir
- quantize_composite_functions_weight_only.mlir
- remove_sharding_custom_call.mlir
- replace_stablehlo_ops_in_main_function_with_xla_call_module_ops.mlir
- restore_function_name.mlir
- shape_cstr_legalize_to_hlo.mlir
- unfuse_mhlo_batch_norm.mlir
- unwrap_xla_call_module_op.mlir
- xla_call_module_to_call.mlir
- process_nchw_tensor.mlir
- BUILD
- fill_quantization_options_test.cc
- stablehlo_quant_opt.cc
- bfloat16_type.cc
- bfloat16_type.h
- bfloat16_type_test.cc
- fill_quantization_options.cc
- fill_quantization_options.h
- math_utils.cc
- math_utils.h
- math_utils_test.cc
- stablehlo_type_utils.h
- stablehlo_type_utils_test.cc
- tf_type_utils.cc
- tf_type_utils.h
- tf_type_utils_test.cc
- BUILD
- internal_visibility_allowlist.bzl
- quantization_config.proto
- quantization_options.proto
- quantize_passes.cc
- quantize_passes.h
- custom_aggregator_op_test.py
- BUILD
- calibration_algorithm.py
- calibration_algorithm_test.py
- calibration_statistics.proto
- calibration_statistics_collector_average_min_max.cc
- calibration_statistics_collector_average_min_max.h
- calibration_statistics_collector_base.h
- calibration_statistics_collector_histogram.cc
- calibration_statistics_collector_histogram.h
- calibration_statistics_collector_min_max.cc
- calibration_statistics_collector_min_max.h
- calibration_statistics_collector_test.cc
- calibration_statistics_saver_op.cc
- calibration_statistics_saver_op_test.cc
- custom_aggregator_op.cc
- pywrap_calibration.pyi
- BUILD
- const_op_size.cc
- const_op_size.h
- const_op_size_test.cc
- constant_fold.cc
- constant_fold.h
- constant_fold_test.cc
- convert_asset_args.cc
- convert_asset_args.h
- convert_asset_args_test.cc
- quantization_unit_loc.cc
- quantization_unit_loc.h
- run_passes.cc
- run_passes.h
- save_variables.cc
- save_variables.h
- save_variables_test.cc
- BUILD
- dump_tensor_op.cc
- mlir_dump.cc
- mlir_dump.h
- mlir_dump_test.cc
- BUILD
- tf_op_quant_spec.cc
- tf_op_quant_spec.h
- tf_op_quant_spec_test.cc
- tf_quantize_op.cc
- tf_quantize_op.h
- tf_quantize_op_test.cc
- uniform_op_quant_spec.cc
- uniform_op_quant_spec.h
- add_dump_tensor_op.cc
- add_quantization_unit_loc.cc
- cast_bf16_ops_to_f32.cc
- cast_bf16_ops_to_f32.td
- constants.h
- convert_custom_aggregation_op_to_quant_stats.cc
- convert_fake_quant_to_qdq.cc
- convert_tf_xla_op_to_tf_op.cc
- convert_tf_xla_op_to_tf_op.td
- convert_tpu_model_to_cpu.cc
- convert_tpu_model_to_cpu.td
- duplicate_shape_determining_constants.cc
- insert_custom_aggregation_ops.cc
- insert_main_function.cc
- insert_quantized_functions.cc
- insert_restore_op.cc
- insert_save_op.cc
- lift_hashtable_ops_as_args.cc
- lift_quantizable_spots_as_functions.cc
- lift_quantizable_spots_as_functions.td
- lift_quantizable_spots_as_functions_drq.cc
- lift_quantizable_spots_as_functions_drq.td
- manipulate_model_attr.cc
- manipulate_model_attr.h
- mark_functions_noinline.cc
- merge_duplicate_resource_ops.cc
- merge_initializer_function_ops_to_main.cc
- merge_save_function_ops_to_main.cc
- optimize.cc
- optimize.td
- passes.h
- post_quantize.cc
- post_quantize.td
- prepare_lifting.cc
- prepare_lifting.td
- prepare_quantize.cc
- prepare_quantize.td
- prepare_quantize_drq.cc
- preprocess_op.cc
- preprocess_op.td
- propagate_quantize_type.cc
- quantize.cc
- quantize_composite_functions.cc
- quantize_composite_functions.td
- quantize_weights.cc
- quantized_function_library.mlir
- quantized_function_library_tf_drq.mlir
- quantized_function_library_uniform_quantized.mlir
- quantized_function_library_uniform_quantized_drq.mlir
- quantized_function_library_xla_weight_only.mlir
- remove_identity_op_pattern.cc
- remove_identity_op_pattern.h
- remove_var_init_by_const.cc
- replace_cast_hacks_with_tf_xla_ops.cc
- replace_cast_hacks_with_tf_xla_ops.td
- tf_quant_ops.cc
- tf_quant_ops.h
- tf_quant_ops.td
- tf_quant_opt.cc
- unfreeze_constants.cc
- concurrency_test.py
- quantize_model_test.py
- quantize_model_test_base.py
- BUILD
- py_function_lib.h
- py_function_lib.py
- py_function_lib_test.py
- pywrap_function_lib.cc
- pywrap_function_lib.pyi
- pywrap_quantize_model.cc
- pywrap_quantize_model.pyi
- pywrap_quantize_model_test.py
- quantize_model.cc
- quantize_model.h
- quantize_model.py
- representative_dataset.py
- representative_dataset_test.py
- save_model.py
- type_casters.h
- unfreeze_constants.cc
- unfreeze_constants.h
- add_dump_tensor_op.mlir
- add_dump_tensor_op_stablehlo.mlir
- add_quantization_unit_loc.mlir
- BUILD
- cast_bf16_ops_to_f32.mlir
- convert_custom_aggregation_op_to_quant_stats.mlir
- convert_fake_quant_to_qdq.mlir
- convert_tf_xla_op_to_tf_op.mlir
- convert_tpu_model_to_cpu.mlir
- duplicate_shape_determining_constants.mlir
- fake_quant_e2e_flow.mlir
- fake_quant_e2e_xla.mlir
- insert_custom_aggregation_ops.mlir
- insert_main_function.mlir
- insert_quantized_functions.mlir
- insert_quantized_functions_drq.mlir
- insert_quantized_functions_weight_only.mlir
- insert_restore_op.mlir
- insert_save_op.mlir
- lift_hashtable_ops_as_args.mlir
- lift_quantizable_spots_as_functions.mlir
- lift_quantizable_spots_as_functions_drq.mlir
- lift_quantizable_spots_as_functions_drq_min_elements.mlir
- lift_quantizable_spots_as_functions_xla.mlir
- lift_quantizable_spots_as_functions_xla_selective_quantization.mlir
- mark_functions_noinline.mlir
- merge_duplicate_resource_ops.mlir
- merge_initializer_function_ops_to_main.mlir
- merge_save_function_ops_to_main.mlir
- optimize.mlir
- prepare_lifting.mlir
- prepare_quantize.mlir
- prepare_quantize_drq.mlir
- prepare_quantize_drq_per_channel.mlir
- prepare_quantize_ptq.mlir
- prepare_quantize_ptq_per_channel.mlir
- preprocess_op.mlir
- preprocess_op_weight_only.mlir
- propagate_quantize_type.mlir
- quantize.mlir
- quantize_composit_functions_debugging.mlir
- quantize_composite_functions.mlir
- quantize_composite_functions_drq.mlir
- quantize_composite_functions_weight_only.mlir
- quantize_composite_functions_xla.mlir
- quantize_drq.mlir
- quantize_weights.mlir
- quantize_xla.mlir
- remove_var_init_by_const.mlir
- replace_cast_hacks_with_tf_xla_ops.mlir
- replace_cast_hacks_with_tf_xla_ops_large_constants.mlir
- unfreeze_constants.mlir
- BUILD
- fake_quant_utils.cc
- fake_quant_utils.h
- tf_quantize_op_utils.cc
- tf_quantize_op_utils.h
- tf_to_uniform_attribute_utils.cc
- tf_to_uniform_attribute_utils.h
- tf_to_uniform_attribute_utils_test.cc
- tf_to_xla_attribute_utils.cc
- tf_to_xla_attribute_utils.h
- tf_to_xla_attribute_utils_test.cc
- BUILD
- exported_model.proto
- gen_quantized_function_library.py
- internal_visibility_allowlist.bzl
- quantization_options.proto
- quantize_passes.cc
- quantize_passes.h
- quantize_preprocess.cc
- quantize_preprocess.h
- fuse_convolution_pass.cc
- fuse_convolution_pass.h
- README.md
- unfuse_batch_norm_pass.cc
- unfuse_batch_norm_pass.h
- fold_broadcast_pass.cc
- fold_broadcast_pass.h
- legalize_tf.cc
- legalize_tf_passes.h
- legalize_tf_patterns.td
- legalize_tf_xla_call_module_to_stablehlo_pass.cc
- legalize_tf_xla_call_module_to_stablehlo_pass.h
- rename_entrypoint_to_main.cc
- rename_entrypoint_to_main.h
- tf_stablehlo_pass.cc
- tf_stablehlo_pass.h
- utils.cc
- utils.h
- utils_test.cc
- BUILD
- stablehlo.cc
- stablehlo.py
- stablehlo_test.py
- per_function_aggregate_analysis.h
- resource_alias_analysis.cc
- resource_alias_analysis.h
- resource_dataflow.cc
- resource_dataflow.h
- resource_value_typed_analyzer.cc
- resource_value_typed_analyzer.h
- side_effect_analysis.cc
- side_effect_analysis.h
- tf_dataflow.h
- BUILD
- c_api_unified_experimental_mlir.cc
- c_api_unified_experimental_mlir_registration.cc
- space_to_depth_transform.png
- enable_mlir_bridge.md
- space_to_depth.md
- tf_dialects.md
- BUILD
- tfrt_ops.cc
- tfrt_ops.h
- tfrt_ops.td
- tf_arith_ops_folder.cc
- tf_arith_ops_folder.h
- tf_attributes.h
- tf_device.cc
- tf_device.h
- tf_device_ops.td
- tf_dialect.h
- tf_executor.cc
- tf_executor.h
- tf_executor_ops.td
- tf_generated_ops.td
- tf_op_base.td
- tf_op_interfaces.cc
- tf_op_interfaces.h
- tf_op_interfaces.td
- tf_ops.cc
- tf_ops.h
- tf_ops.td
- tf_ops_a_m.cc
- tf_ops_a_m.h
- tf_ops_canonicalization_helper.h
- tf_ops_device_helper.cc
- tf_ops_device_helper.h
- tf_ops_layout_helper.cc
- tf_ops_layout_helper.h
- tf_ops_n_z.cc
- tf_ops_n_z.h
- tf_ops_tensor_helper.cc
- tf_ops_tensor_helper.h
- tf_remaining_ops.cc
- tf_remaining_ops.h
- tf_saved_model.cc
- tf_saved_model.h
- tf_saved_model_ops.td
- tf_saved_model_test.cc
- tf_side_effects.h
- tf_structs.cc
- tf_structs.h
- tf_traits.h
- tf_types.def
- tf_types.h
- tf_verifiers.cc
- tf_verifiers.h
- tpu_embedding_ops_registry.cc
- tpu_embedding_ops_registry.h
- mlir_local_var_op.cc
- mlir_passthrough_op.cc
- add.mlir
- argument-sharding-invalid.mlir
- argument-sharding.mlir
- BUILD
- constant-folding-hook.mlir
- constant-folding.mlir
- convert_mhlo_quant_to_int.mlir
- graph-resource.mlir
- graph-resource.pbtxt
- graph.pbtxt
- mlir-module-serialized-str-attr.mlir
- replicate-tensor-list-init-ops.mlir
- result-sharding.mlir
- serialized-mlir-module-str-attr-invalid.mlir
- serialized-mlir-module-str-attr.mlir
- shape-inference-after-legalization.mlir
- shape-inference.mlir
- stablehlo_add.mlir
- BUILD
- executor_tpuv1_island_coarsening.mlir
- while_op.mlir
- BUILD
- executor_tpuv1_inline_tpu_island.mlir
- while_op.mlir
- BUILD
- case_op.mlir
- executor_tpuv1_outline_tpu_island.mlir
- while_op.mlir
- BUILD
- saved_model.pbtxt
- add.pbtxt
- arg-as-fetch.pbtxt
- arg-control-dep.pbtxt
- arg-data-type-with-subtype.pbtxt
- arg-data-type.pbtxt
- arg-multi-data-type-with-subtype.pbtxt
- arg-retval-attrs.pbtxt
- BUILD
- case_op.pbtxt
- const-values.pbtxt
- device-arg-retval-attr.pbtxt
- empty-input-shapes.pbtxt
- empty-value-attr.pbtxt
- error-message-with-source-info.pbtxt
- error-message-with-source-info.pbtxt.debug
- error-message-with-source-info.pbtxt.fake_py.debug
- feed-as-fetch.pbtxt
- feed-control-dep.pbtxt
- force_shared_name_for_resource_ops.pbtxt
- function-func-attr.pbtxt
- functional-if-ops.pbtxt
- functional-while-ops.pbtxt
- graph-as-function-control-ret.pbtxt
- graph-as-function-retval-of-arg.pbtxt
- graph-as-function.pbtxt
- graph-custom-operation.pbtxt
- graph-default-attr.pbtxt
- graph-device-retval.pbtxt
- graph-empty-tensor-content.pbtxt
- graph-func-attr.pbtxt
- graph-function-call.pbtxt
- graph-function-control-ret-diff-island.pbtxt
- graph-function-control-ret-same-island.pbtxt
- graph-function-defs.pbtxt
- graph-function-input-shapes.pbtxt
- graph-function-name-bug.pbtxt
- graph-function-resource-args.pbtxt
- graph-gradient-def.pbtxt
- graph-input-func-arg-name-collision.pbtxt
- graph-library.pbtxt
- graph-malformed.pbtxt
- graph-scalar-input.pbtxt
- graph-uint8-return.pbtxt
- graph-undefined-output.pbtxt
- graph-version-info.pbtxt
- graph-while-loop.pbtxt
- invalid-output-index.pbtxt
- legacy-fed-input-without-inputs.pbtxt
- merge_node_with_function.pbtxt
- mlir_passthrough_op.pbtxt
- multi-output-feeds.pbtxt
- multiple-use-next-iteration.pbtxt
- node-locations.pbtxt
- node-order.pbtxt
- output-shapes-attr.pbtxt
- output-shapes.pbtxt
- parse_example.pbtxt
- parse_example_v2.pbtxt
- partial-device-name.pbtxt
- prune_unused_nodes.pbtxt
- quint8-const.pbtxt
- shape-attrs.pbtxt
- stateful-attribute.pbtxt
- string-attr.pbtxt
- switch_n.pbtxt
- target.pbtxt
- tensor-list.pbtxt
- tf-data-pipeline.pbtxt
- unregistered_kernel.pbtxt
- BUILD
- convert_tensor.mlir
- aliasing_arg_attr.mlir
- BUILD
- case.mlir
- convert_tensor.mlir
- derived_shape_attr.mlir
- derived_size_attr.mlir
- device-arg-retval-attr.mlir
- export_main_to_flib.mlir
- fetch_feed_names.mlir
- func_attr.mlir
- func_list_attr.mlir
- function-control-ret.mlir
- function-order.mlir
- function-resource-args-handle-info.mlir
- function-resource-args.mlir
- functional-if-ops.mlir
- functional-while-ops.mlir
- graph-as-function.mlir
- infer_derived_attribute.mlir
- invalid_input.mlir
- legalized_name.mlir
- missing-main.mlir
- noop.mlir
- optional_symbol_ref.mlir
- output-shapes-attr.mlir
- parse_example.mlir
- parse_example_v2.mlir
- preserve-entry-func-names.mlir
- ref-type-attr.mlir
- ref-while-loop.mlir
- shape_list_attr.mlir
- simple.mlir
- stringescape.mlir
- switchn.mlir
- tf-gradient-attr.mlir
- tf-legacy-call.mlir
- tf_add.mlir
- tf_identity_n.mlir
- tf_tpu_embedding_ops.mlir
- type_attr.mlir
- type_list_attr.mlir
- unique_name.mlir
- unique_output_name.mlir
- while-loop.mlir
- unsupported-op-test.sh
- add_functions_for_exported_names.mlir
- annotate-parameter-replication.mlir
- batchmatmul_to_einsum.mlir
- breakup-islands.mlir
- BUILD
- cannonicalize_ops_outside_compilation.mlir
- canonicalize.mlir
- canonicalize_compile_and_replicate_attributes.mlir
- check_control_dependencies.mlir
- cluster_formation.mlir
- cluster_ops_by_policy.mlir
- cluster_outlining.mlir
- cluster_tf_ops_pass.mlir
- colocate_tpu_copy_with_dynamic_shape.mlir
- constant-fold.mlir
- constant_op_device_assignment.mlir
- convert-tf-control-flow-to-scf.mlir
- convert_control_to_data_outputs.mlir
- convert_launch_func_to_tf_call.mlir
- convert_session_initializer_to_function.mlir
- convert_to_legacy_compile_and_replicate_attributes.mlir
- decompose_optionals.mlir
- decompose_reduce_dataset.mlir
- decompose_resource_ops.mlir
- device_assignment.mlir
- device_assignment_by_func_attr.mlir
- device_attribute_to_launch.mlir
- device_canonicalize.mlir
- device_copy.mlir
- drop_while_shape_invariant.mlir
- einsum.mlir
- embedding_pipelining.mlir
- embedding_program_key.mlir
- embedding_sequencing.mlir
- empty-main.mlir
- end-to-end-tpu-reshard-variables.mlir
- executor_canonicalize.mlir
- executor_island_coarsening.mlir
- executor_island_materialize_const.mlir
- extract_head_tail_outside_compilation.mlir
- extract_outside_compilation.mlir
- extract_tpu_copy_with_dynamic_shape_op.mlir
- fold-broadcast.mlir
- freeze_variables.mlir
- func-attr-invalid.mlir
- func-attr.mlir
- functional-control-flow-to-cfg.mlir
- functional-control-flow-to-regions.mlir
- functionalize-if-fail.mlir
- functionalize-if.mlir
- fused_kernel_matcher.mlir
- gpu_fusion.mlir
- graph_pruning.mlir
- graph_pruning_preserve_ops.mlir
- group_by_dialect.mlir
- guarantee-all-funcs-one-use.mlir
- hoist_broadcast_read.mlir
- hoist_loop_invariant.mlir
- hoist_replicate_invariant_resource_writes.mlir
- ici_weight_distribution_non_spmd_clustering_end_to_end.mlir
- ici_weight_distribution_non_spmd_mlir_end_to_end.mlir
- ici_weight_distribution_spmd_clustering_end_to_end.mlir
- ici_weight_distribution_spmd_mlir_end_to_end.mlir
- init_text_file_to_import.mlir
- init_text_file_to_import_invalid.mlir
- init_text_file_to_import_saved_model.mlir
- inlining.mlir
- isolate-placer.mlir
- launch_outlining.mlir
- launch_to_device_attribute.mlir
- launch_to_device_attribute_legacy.mlir
- layout_optimization_layout_assignment_gpu_cc_60.mlir
- layout_optimization_layout_assignment_gpu_cc_70.mlir
- layout_optimization_layout_assignment_to_nchw.mlir
- layout_optimization_layout_assignment_to_nhwc.mlir
- layout_optimization_move_transposes_begin.mlir
- layout_optimization_move_transposes_end.mlir
- layout_optimization_to_nchw.mlir
- layout_optimization_to_nhwc.mlir
- legalize_tfg.mlir
- legalize_tfg_arg_control_dep.mlir
- legalize_tfg_with_control_flow.mlir
- localize_var_handles.mlir
- lower_globals_to_ml_program.mlir
- lower_globals_to_ml_program_invalid.mlir
- lower_quantized.mlir
- lower_tf.mlir
- lower_variable_ops_to_ml_program.mlir
- mark_input_output_aliases.mlir
- mark_ops_for_outside_compilation.mlir
- materialize_passthrough_op.mlir
- merge_control_flow.mlir
- mlprogram.mlir
- move_tpu_compile_to_front.mlir
- name_anonymous_iterators.mlir
- optimize-arg-operand-constraint.mlir
- optimize.mlir
- order_by_dialect.mlir
- order_for_program_key.mlir
- parallel_execute_to_islands.mlir
- parallel_execute_to_islands_legacy.mlir
- prepare_tpu_computation_for_tf_export.mlir
- print.mlir
- promote_resources_to_args.mlir
- promote_resources_to_args_functions.mlir
- promote_var_handles_to_args.mlir
- readonly_references_to_resources.mlir
- region-control-flow-to-functional.mlir
- remove_unused_arguments.mlir
- remove_unused_while_results.mlir
- replica_id_to_device_ordinal.mlir
- replicate_invariant_op_hoisting.mlir
- replicate_tensor_list_init_ops.mlir
- replicate_to_island.mlir
- replicate_to_island_legacy.mlir
- resource-alias-analysis-test.mlir
- resource-device-inference.mlir
- resource_analyzer.mlir
- resource_inlining.mlir
- resource_op_lifting.mlir
- rewrite_tpu_embedding_ops.mlir
- shape_inference.mlir
- shape_inference_for_xla_call_module.mlir
- shape_inference_with_shape_specialization.mlir
- side-effect-analysis-test.mlir
- sink_constant.mlir
- split_into_island_per_op.mlir
- stack_ops_decomposition.mlir
- strip_noinline.mlir
- strip_saved_module_metadata.mlir
- strip_tf_attributes.mlir
- tensor_array_ops_decomposition.mlir
- tensor_list_ops_decomposition.mlir
- tf-executor-to-functional.mlir
- tf-functional-to-executor.mlir
- tf-ops.mlir
- tf-reduce-identity.mlir
- BUILD
- dialect_registration.h
- BUILD
- glob_lit_test.bzl
- init_mlir.cc
- init_mlir.h
- mlir_graph_optimization_pass.cc
- mlir_graph_optimization_pass.h
- mlir_graph_optimization_pass_registration.cc
- mlir_graph_optimization_pass_test.cc
- op_or_arg_name_mapper.cc
- op_or_arg_name_mapper.h
- op_or_arg_name_mapper_test.cc
- README.md
- register_common_dialects.cc
- register_common_dialects.h
- register_common_dialects_test.cc
- runlit.cfg.py
- runlit.site.cfg.py
- OWNERS_FORCE_SUBMIT_APPROVERS
- .clang-format
- __init__.py
- api_template.__init__.py
- api_template_v1.__init__.py
- BUILD
- build_cleaner_spec.textproto
- compat_template.__init__.py
- compat_template_v1.__init__.py
- .bazelignore
- .bazelrc
- .bazelversion
- .clang-format
- .gitignore
- .pylintrc
- .zenodo.json
- arm_compiler.BUILD
- AUTHORS
- bazel_downloader.cfg
- BUILD
- CITATION.cff
- CODE_OF_CONDUCT.md
- CODEOWNERS
- configure
- configure.cmd
- configure.py
- CONTRIBUTING.md
- ISSUES.md
- LICENSE
- models.BUILD
- MODULE.bazel
- README.md
- RELEASE.md
- requirements_lock_3_10.txt
- requirements_lock_3_11.txt
- requirements_lock_3_12.txt
- requirements_lock_3_13.txt
- requirements_lock_3_14.txt
- SECURITY.md
- WORKSPACE
# 설치 가이드
1. 코드 내려받기
git clone https://github.com/tensorflow/tensorflow
깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.
cd tensorflow
방금 내려받은 프로젝트 폴더 안으로 이동합니다.
2. 공식 설치 스크립트
쉬움 추천사전 준비물
- Python 3 pip 명령어를 쓰려면 Python이 필요합니다.
pip install tensorflow
PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.
pip install tensorflow-cpu
PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.
설치 후 새 터미널을 열고, 프로그램의 버전 확인 명령(예: --version)으로 정상 설치됐는지 확인하세요.
이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.
3. Docker
쉬움사전 준비물
- Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
- Docker Desktop 컨테이너를 빌드하고 실행하려면 필요합니다. 설치 후 실행해서 백그라운드에 켜두세요.
⚠️ 이 프로젝트는 규모가 큰 저장소라, 이 방법이 실제 핵심 제품이 아니라 내부 하위 패키지를 가리키는 것일 수 있습니다. README 전체를 함께 확인해보세요.
docker build -f ci/devinfra/docker/windows/Dockerfile -t tensorflow .
Dockerfile을 기반으로 실행 가능한 이미지를 빌드합니다.
docker run -p 8080:80 tensorflow
빌드된 이미지를 실제 컨테이너로 실행합니다.
터미널에 docker compose ps 를 입력해 컨테이너들이 Up 상태인지 확인하세요. README에 포트 번호가 적혀있다면 브라우저에서 http://localhost:포트번호 로 접속해보세요.
// repository documentation
Was this content helpful?
(0 ratings)