MCPcopy Create free account

hub / github.com/alibaba/BladeDISC / functions

Functions6,088 in github.com/alibaba/BladeDISC

↓ 1 callersMethodmergeEdgePaddingHigh
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:1075
↓ 1 callersMethodmergeEdgePaddingLow
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:1063
↓ 1 callersMethodmergeInteriorPadding
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:1087
↓ 1 callersMethodmergeLimitIndex
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:963
↓ 1 callersFunctionmergeShards
tao_compiler/mlir/xla/ral/context/custom_library/tf_topk.cu.h:249
↓ 1 callersFunctionmergeSkeletonGroupsInOrder
tao_compiler/mlir/disc/transforms/fusion_utils.cc:985
↓ 1 callersMethodmergeStartIndex
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:951
↓ 1 callersMethodmergeStride
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:981
↓ 1 callersMethodmergeWithoutInit
Merges two fusion patterns and returns the merged pattern. The original pattern remains unmodified. The new merged pattern is uninitialized.
tao_compiler/mlir/disc/transforms/fusion_utils.cc:785
↓ 1 callersFunctionmin_group_nodes
(min_num_nodes)
pytorch_blade/torch_blade/clustering/support_fusion_group.py:23
↓ 1 callersFunctionmkl_install_dir
(root)
scripts/python/common_setup.py:258
↓ 1 callersFunctionmkl_install_dir
(root)
pytorch_blade/common_setup.py:258
↓ 1 callersFunctionmkl_ral_batch_gemm
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.cc:839
↓ 1 callersFunctionmkl_ral_gemm
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.cc:628
↓ 1 callersFunctionmod_ptr
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/utils.hpp:270
↓ 1 callersMethodnameNewOperation
tao_compiler/mlir/disc/tools/disc-transform/utils.cc:32
↓ 1 callersFunctionneed_preserve
( simple_graph: SimpleGraph, node_names: Union[str, List[str]], pattern_map: Optional[Dict[str, st
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:885
↓ 1 callersMethodnelems
returns the number of elements including padding if \param with_padding * is true, and the number of data elements otherwise */
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:131
↓ 1 callersFunctionnewBcastOp
tao_compiler/mlir/disc/transforms/disc_algebraic_simplifier.cc:263
↓ 1 callersFunctionnn_ops_first_input_preserving
Requirements: dims : preserved from the first argument scalar type : preserved from the first argument (doesn't have to match other argum
pytorch_blade/pytorch_blade/compiler/jit/torch/op_registry.cpp:32
↓ 1 callersMethodnodes_by_function_type
(self, function_types: Iterable[Callable])
tools/torch_quant/torch_quant/graph.py:148
↓ 1 callersMethodnodes_by_method_names
(self, method_names: Iterable[str])
tools/torch_quant/torch_quant/graph.py:155
↓ 1 callersMethodnon_negitive_output
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/attributes.hpp:91
↓ 1 callersFunctionnormal_init_
(weights)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/ops.py:82
↓ 1 callersMethodnum_constant_args
tao/tao_bridge/executable.h:175
↓ 1 callersMethodnum_groups
(self)
pytorch_blade/torch_blade/clustering/support_fusion_algorithm.py:52
↓ 1 callersMethodnum_sets
(self)
pytorch_blade/torch_blade/algorithm/union_set.py:38
↓ 1 callersMethodonContextFinish
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:189
↓ 1 callersMethodonExecutionFinish
tao_compiler/mlir/xla/ral/context/tensorflow/tf_context_impl.cc:102
↓ 1 callersMethodonExecutionStart
tao_compiler/mlir/xla/ral/ral_context.cc:148
↓ 1 callersFunctiononednn_ral_batch_gemm
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.cc:868
↓ 1 callersFunctiononednn_ral_gemm
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.cc:700
↓ 1 callersMethodoperand
tao/tao_bridge/tf/deadness_analysis.cc:222
↓ 1 callersMethodoperator()
tao/tao_bridge/passes/functionalize_cond.cc:104
↓ 1 callersFunctionopt_disc_config
()
examples/PyTorch/Inference/CPU/albert/main.py:43
↓ 1 callersFunctionoptimize
(model, inputs, optimize_config)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/foldacc.py:73
↓ 1 callersFunctionoptimizeShapeComputation
It’s an iterative process and each iteration can be divided into five steps: - Step #1: canonicalization. including normal canonicalization pattern, c
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:1775
↓ 1 callersMethodoptimize_graph_def
( self, graph_def: tf.GraphDef, protected_nodes: List[str] )
tensorflow_blade/tf_blade/common/tf_grappler.py:25
↓ 1 callersFunctionoptimize_openfold
( model, model_config, enable_disc=True, enable_low_mem=False, enable_amp=True, ena
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/fold/openfold.py:18
↓ 1 callersFunctionoptimize_unifold
( model, model_config, enable_disc=True, enable_low_mem=False, enable_amp=True, ena
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/fold/unifold.py:69
↓ 1 callersFunctionoutlineKernelFunc
tao_compiler/mlir/disc/transforms/disc_outline_cpu_kernel.cc:184
↓ 1 callersFunctionoutlineKernelFuncImpl
Outline the `launch` operation body into a kernel function. Replace `scf.yield` operations by `return` in the generated function.
tao_compiler/mlir/disc/transforms/disc_outline_cpu_kernel.cc:119
↓ 1 callersMethodowned
Returns true if this wrapper is the exclusive owner
tao_compiler/mlir/xla/ral/context/base/base_context.h:57
↓ 1 callersMethodpacked_weight
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.cc:589
↓ 1 callersFunctionparseArrayAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:120
↓ 1 callersFunctionparseBoolAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:58
↓ 1 callersFunctionparseCommonStructuredOpParts
Common parsing used for both named structured ops created by ods-gen and by manually defined C++ ops. Does not handle regions.
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:345
↓ 1 callersFunctionparseConvCustomeAttr
tao_compiler/mlir/xla/ral/context/common_context_impl_quantization.h:27
↓ 1 callersFunctionparseConvParams
tao_compiler/mlir/disc/transforms/lower_tf.cc:928
↓ 1 callersFunctionparseDenseElementsAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:82
↓ 1 callersFunctionparseDictAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:34
↓ 1 callersFunctionparseFloatAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:74
↓ 1 callersFunctionparseIntArrayAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:134
↓ 1 callersFunctionparseIntAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:66
↓ 1 callersFunctionparseMLIRInput
tao_compiler/mlir/disc/disc_compiler_main.cc:91
↓ 1 callersFunctionparseMLIRInput
tao_compiler/mlir/disc/tools/disc-source-emitter/disc-cuda-emitter.cc:37
↓ 1 callersFunctionparseMLIRInput
tao_compiler/mlir/disc/tools/disc-pdll/disc-pdll.cc:52
↓ 1 callersFunctionparseMLIRInput
pytorch_blade/tests/torch-disc-pdll/torch-disc-pdll.cpp:43
↓ 1 callersMethodparseModuleFromFile
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:1509
↓ 1 callersFunctionparseNamedStructuredOpResults
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:400
↓ 1 callersFunctionparseShape
tao_compiler/mlir/xla/ral/test/raw_cuda_test.cc:60
↓ 1 callersFunctionparseStrAttr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:50
↓ 1 callersFunctionparse_args
()
scripts/python/tao_build.py:693
↓ 1 callersFunctionparse_args
()
scripts/python/common_setup.py:742
↓ 1 callersFunctionparse_args
()
examples/PyTorch/Train/Lazy/Bert/bert_train_ltc.py:185
↓ 1 callersFunctionparse_args
()
examples/PyTorch/Train/Dynamo/Bert/test_bert.py:202
↓ 1 callersFunctionparse_args
()
tensorflow_blade/build.py:269
↓ 1 callersFunctionparse_args
()
pytorch_blade/common_setup.py:742
↓ 1 callersFunctionparse_envs
tao_compiler/decoupling/tao_compiler_main.cc:51
↓ 1 callersFunctionparse_fasta
(fasta_string)
examples/PyTorch/Inference/CUDA/AlphaFold/openfold_inference.py:42
↓ 1 callersFunctionparse_graphs
pytorch_blade/tests/torchscript/shape_analysis_tool.cpp:49
↓ 1 callersFunctionparse_version
Parses a version string into (major, minor, patch) version numbers. Args: version: Full version number string, possibly including revi
pytorch_blade/torch_blade/utils.py:207
↓ 1 callersFunctionpatternKindFromString
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:1385
↓ 1 callersFunctionpatternKindToString
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:1377
↓ 1 callersMethodplatform_id
This is equal to xla_device_metadata()->platform()->id() if xla_device_metadata() is not nullptr.
tao/tao_bridge/kernels/platform_info.h:41
↓ 1 callersFunctionpopulateDiscAlgebraicSimplifierPatterns
tao_compiler/mlir/disc/transforms/disc_algebraic_simplifier.cc:543
↓ 1 callersFunctionpopulateDiscDenseToSparsePatterns
tao_compiler/mlir/disc/transforms/disc_dense_to_sparse.cc:395
↓ 1 callersFunctionpopulateDiscHLOToLHLOConversionPattern
tao_compiler/mlir/disc/transforms/disc_hlo_legalize_to_lhlo.cc:270
↓ 1 callersFunctionpopulateDiscSparseGemmTransposeSimplifierPatterns
tao_compiler/mlir/disc/transforms/disc_dense_to_sparse.cc:518
↓ 1 callersFunctionpopulateDiscToLLVMConversionPatterns
tao_compiler/mlir/disc/transforms/disc_to_llvm.cc:1338
↓ 1 callersFunctionpopulateGpuQuantizeAndDequantizePatterns
tao_compiler/mlir/disc/transforms/disc_lower_quantize_and_dequantize.cc:214
↓ 1 callersFunctionpopulateGpuToROCDLConversionPatterns
tao_compiler/mlir/disc/transforms/disc_lower_gpu_ops_to_rocdl_ops.cc:140
↓ 1 callersFunctionpopulateQuantizeAndDequantizePatterns
tao_compiler/mlir/disc/transforms/disc_lower_quantize_and_dequantize.cc:141
↓ 1 callersFunctionpopulateQuantizedPatterns
tao_compiler/mlir/disc/transforms/disc_convert_fake_quant_op.cc:272
↓ 1 callersFunctionpopulateShapeOptimizationPatterns
Adds shape optimization related patterns.
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:631
↓ 1 callersFunctionpopulateTransposeSimplifierPatterns
tao_compiler/mlir/disc/transforms/disc_transpose_simplifier.cc:205
↓ 1 callersMethodprepare
(self)
pytorch_blade/torch_blade/quantization/prepare_data.py:209
↓ 1 callersFunctionprepare_env
Prepare env vars related to building .
scripts/python/tao_build.py:552
↓ 1 callersFunctionprepare_for_data_collect
(model)
pytorch_blade/tests/quantization/test_data_prepare.py:26
↓ 1 callersFunctionprintCommonStructuredOpParts
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:386
↓ 1 callersFunctionprintErrorIfAny
tao_compiler/mlir/disc/tests/mlir_test.cc:162
↓ 1 callersFunctionprintNamedStructuredOpResults
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:394
↓ 1 callersFunctionprintStats
(backend, timings, precision)
examples/PyTorch/Train/Lazy/Bert/bert_train_ltc.py:46
↓ 1 callersFunctionprintStats
(backend, timings, precision)
examples/PyTorch/Train/Dynamo/Bert/test_bert.py:56
↓ 1 callersFunctionprintStats
(backend, timings, precision, batch_size=1)
examples/PyTorch/Inference/CUDA/ResNet/main.py:28
↓ 1 callersFunctionprintStats
(func_name, backend, timings, precision, batch_size=1)
pytorch_blade/benchmark/detectron2/run_blade.py:30
↓ 1 callersFunctionprint_checks
pytorch_blade/tests/torchscript/shape_analysis_tool.cpp:90
↓ 1 callersFunctionprint_output_shape
tao_compiler/mlir/xla/ral/test/raw_cuda_test.cc:87
↓ 1 callersFunctionprint_output_shape
tao_compiler/mlir/disc/tests/mlir_test.cc:182
↓ 1 callersFunctionpromoteConv1DToConv2D
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.cc:130
← previousnext →2,401–2,500 of 6,088, ranked by callers