File tree

• .circleci
  • cimodel/data
    • pytorch_build_definitions.py
  • config.yml
  • docker
    • build.sh
  • verbatim-sources
    • job-specs-custom.yml
    • workflows-docker-builder.yml
    • workflows-ecr-gc.yml
    • workflows-nightly-android-binary-builds.yml
    • workflows-pytorch-ge-config-tests.yml
    • workflows-pytorch-mobile-builds.yml
• .jenkins
  • caffe2
    • test.sh
  • pytorch
    • common.sh
• BUILD.bazel
• CMakeLists.txt
• CONTRIBUTING.md
• WORKSPACE
• aten/src
  • ATen
    • CPUGeneratorImpl.cpp
    • Declarations.cwrap
    • SparseTensorImpl.cpp
    • Utils.h
    • autocast_mode.cpp
    • core
      • Generator.h
      • LegacyTypeDispatch.h
      • VariableFallbackKernel.cpp
      • aten_interned_strings.h
      • boxing/impl
        • kernel_function_legacy_test.cpp
        • kernel_function_test.cpp
        • kernel_lambda_legacy_test.cpp
        • kernel_lambda_test.cpp
        • kernel_stackbased_test.cpp
        • make_boxed_from_unboxed_functor_test.cpp
      • dispatch
        • DispatchKeyExtractor.h
        • DispatchTable.h
        • Dispatcher.cpp
        • Dispatcher.h
        • OperatorEntry.cpp
        • OperatorEntry.h
      • op_registration
        • README.md
        • op_registration.h
        • op_registration_test.cpp
    • cpu
      • vec256
        • vec256_base.h
        • vec256_bfloat16.h
      • vml.h
    • cuda
      • CUDAGeneratorImpl.cpp
      • CuSparseHandlePool.cpp
      • CublasHandlePool.cpp
      • detail
        • DeviceThreadHandles.h
    • cudnn
      • Handle.cpp
    • function_wrapper.py
    • native_parse.py
    • native
      • Activation.cpp
      • Copy.cpp
      • DistributionTemplates.h
      • Distributions.cpp
      • Math.h
      • Resize.cpp
      • TensorAdvancedIndexing.cpp
      • TensorAdvancedIndexing.h
      • TensorFactories.cpp
      • TensorShape.cpp
      • UnaryOps.cpp
      • UnaryOps.h
      • cpu
        • DistributionTemplates.h
        • MultinomialKernel.cpp
        • UnaryOpsKernel.cpp
      • cuda
        • DistributionBernoulli.cu
        • DistributionCauchyKernel.cu
        • DistributionExponentialKernel.cu
        • DistributionGeometricKernel.cu
        • DistributionLogNormalKernel.cu
        • DistributionNormal.cu
        • DistributionRandomKernel.cu
        • DistributionTemplates.h
        • DistributionUniform.cu
        • Distributions.cu
        • Dropout.cu
        • IndexKernel.cu
        • LegacyDefinitions.cpp
        • MultinomialKernel.cu
        • Normalization.cuh
        • Reduce.cu
        • Resize.cu
        • TensorFactories.cu
      • mkldnn
        • MKLDNNCommon.cpp
      • native_functions.yaml
      • quantized
        • README.md
        • cpu
          • fbgemm_utils.cpp
          • qadd.cpp
          • qbatch_norm.cpp
          • qclamp.cpp
          • qconcat.cpp
          • qconv.cpp
          • qconv_prepack.cpp
          • qconv_unpack.cpp
          • qhardswish.cpp
          • qlinear.cpp
          • qlinear_dynamic.cpp
          • qlinear_prepack.cpp
          • qlinear_unpack.cpp
          • qmul.cpp
          • qnnpack
            • CMakeLists.txt
            • README.md
            • bench
              • hardswish.cc
            • configure.py
            • include
              • pytorch_qnnpack.h
            • src
              • hardswish.c
            • test
              • hardswish-operator-tester.h
              • hardswish.cc
          • qpool.cpp
          • qrelu.cpp
          • tensor_operators.cpp
      • sparse
        • SparseTensor.cpp
      • xnnpack
        • Factory.cpp
        • RegisterOpContextClass.cpp
    • nn.yaml
    • nn_parse.py
    • quantized
      • Quantizer.cpp
    • templates
      • BackendSelectRegister.cpp
      • TensorBody.h
    • test
      • backend_fallback_test.cpp
      • basic.cpp
      • cpu_rng_test.cpp
      • extension_backend_test.cpp
      • pow_test.cpp
      • rng_test.h
      • xla_tensor_test.cpp
  • THCUNN/generic
    • RReLU.cu
    • THCUNN.h
  • THC/generic
    • THCTensor.cpp
    • THCTensorMasked.cu
    • THCTensorMasked.h
    • THCTensorRandom.cu
    • THCTensorRandom.h
  • TH
    • THGeneral.h.in
    • generic
      • THTensor.cpp
      • THTensorRandom.cpp
      • THTensorRandom.h
• benchmarks/operator_benchmark/pt
  • qbatchnorm_test.py
• c10
  • core
    • Backend.h
    • Device.cpp
    • DispatchKey.cpp
    • DispatchKey.h
    • DispatchKeySet.h
    • TensorImpl.h
    • TensorOptions.h
    • impl
      • LocalDispatchKeySet.cpp
      • LocalDispatchKeySet.h
  • util
    • BFloat16-inl.h
    • BFloat16-math.h
    • either.h
• caffe2
  • contrib
    • CMakeLists.txt
    • fakelowp
      • CMakeLists.txt
      • test
        • README.md
        • test_batchmatmul_nnpi_fp16.py
        • test_batchnorm_nnpi_fp16.py
        • test_fc_nnpi_fp16.py
        • test_op_nnpi_fp16.py
        • test_sls_4bit_nnpi_fp16.py
        • test_sls_nnpi_fp16.py
  • core
    • blob_test.cc
    • event.h
    • export_caffe2_op_to_c10.h
    • net_async_base.cc
    • net_async_base.h
    • net_async_scheduling.cc
    • net_test.cc
    • tensor.h
  • operators
    • dataset_ops.cc
    • dataset_ops.h
  • opt
    • bound_shape_inference_test.cc
    • bound_shape_inferencer.cc
    • bound_shape_inferencer.h
  • python
    • fakelowp
      • __init__.py
      • init_shared_libs.py
    • operator_test
      • dataset_ops_test.py
      • storm_test.py
    • optimizer.py
    • optimizer_test.py
  • sgd
    • storm_op.cc
    • storm_op.h
• cmake
  • Dependencies.cmake
  • Summary.cmake
• docs/source
  • torch.rst
• test
  • backward_compatibility
    • check_backward_compatibility.py
  • cpp_extensions
    • msnpu_extension.cpp
    • rng_extension.cpp
  • cpp/tensorexpr
    • gtest.cpp
    • gtest_assert_float_eq.h
    • test_aten.cpp
    • test_loopnest.cpp
  • distributed
    • test_data_parallel.py
    • test_distributed.py
  • jit
    • test_freezing.py
  • mobile/op_deps
    • expected_deps.yaml
    • utils.cpp
    • utils.h
  • onnx
    • expect
      • TestOperators.test_expand.expect
    • test_pytorch_onnx_onnxruntime.py
    • test_utility_funs.py
  • quantization
    • test_numeric_suite.py
    • test_quantized.py
    • test_quantized_nn_mods.py
  • run_test.py
  • test_autograd.py
  • test_cpp_extensions_jit.py
  • test_dataloader.py
  • test_dispatch.py
  • test_jit.py
  • test_mkldnn.py
  • test_nn.py
  • test_optim.py
  • test_throughput_benchmark.py
  • test_torch.py
  • test_type_promotion.py
• third_party
  • asmjit.BUILD
  • fbgemm.BUILD
  • fbgemm
• tools
  • autograd
    • gen_python_functions.py
    • templates
      • python_torch_functions.cpp
  • build_variables.bzl
  • clang_format.py
  • code_analyzer
    • gen_op_registration_whitelist.py
    • op_dependency.cpp
    • run_analyzer.sh
  • jit
    • gen_jit_dispatch.py
  • pyi
    • gen_pyi.py
• torch
  • CMakeLists.txt
  • _overrides.py
  • _torch_docs.py
  • csrc
    • api/src/optim
      • lbfgs.cpp
    • autograd
      • VariableTypeManual.cpp
      • engine.cpp
      • engine.h
    • distributed
      • autograd
        • context
          • container.cpp
          • container.h
          • context.cpp
        • engine
          • dist_engine.cpp
      • rpc
        • process_group_agent.cpp
        • python_functions.cpp
        • request_callback_impl.cpp
        • rpc_agent.cpp
        • rref_context.cpp
        • rref_context.h
        • torchscript_functions.cpp
    • jit
      • frontend
        • tracer.cpp
        • tracer.h
      • ir
        • ir.cpp
        • ir.h
      • mobile
        • register_mobile_ops.cpp
      • passes
        • freeze_module.cpp
        • onnx
          • constant_fold.cpp
          • function_substitution.cpp
          • function_substitution.h
      • python
        • init.cpp
        • python_tracer.cpp
        • python_tracer.h
        • script_init.cpp
      • runtime
        • interpreter.cpp
        • operator.h
        • register_prim_ops.cpp
        • register_prim_ops_fulljit.cpp
      • tensorexpr
        • loopnest.cpp
        • loopnest.h
    • utils
      • future.h
      • init.cpp
      • python_arg_parser.h
      • python_dispatch.cpp
      • tensor_new.cpp
      • throughput_benchmark-inl.h
      • throughput_benchmark.h
  • distributed
    • distributed_c10d.py
  • functional.py
  • jit
    • __init__.py
  • nn
    • modules
      • _functions.py
      • module.py
    • parallel
      • replicate.py
    • quantized
      • dynamic/modules
        • rnn.py
      • modules
        • batchnorm.py
        • conv.py
        • utils.py
  • onnx
    • utils.py
  • quantization
    • _numeric_suite.py
  • quasirandom.py
  • testing/_internal
    • common_distributed.py
    • common_utils.py
  • utils
    • cpp_extension.py
    • data
      • dataloader.py
    • hipify
      • hipify_python.py
    • tensorboard
      • _embedding.py
      • summary.py
      • writer.py
    • throughput_benchmark.py