MCPcopy Create free account

hub / github.com/alibaba/BladeDISC / functions

Functions6,088 in github.com/alibaba/BladeDISC

↓ 4 callersFunctionMakeBaseCpuContext
tao_compiler/mlir/xla/ral/context/base/cpu/cpu_context_impl.cc:67
↓ 4 callersMethodMerge
Merge two symbolDim if they are compatible.
tao_compiler/mlir/disc/transforms/shape_utils.cc:45
↓ 4 callersFunctionNDef
Helper to construct a NodeDef.
tao/tao_bridge/tao_util_test.cc:66
↓ 4 callersFunctionNew64
tao_compiler/mlir/xla/ral/context/random_impl.cc:42
↓ 4 callersFunctionParseFileString
tao_compiler/mlir/disc/transforms/disc_pdl_utils.cc:445
↓ 4 callersFunctionPrintEdges
tao/tao_bridge/tf/graphcycles_test.cc:80
↓ 4 callersFunctionPrintGCEdges
tao/tao_bridge/tf/graphcycles_test.cc:90
↓ 4 callersMethodRegisterBackend
static */
tao/tao_bridge/tf/xla_op_registry.cc:138
↓ 4 callersMethodRemoveEdge
tao_compiler/mlir/disc/utils/cycle_detector.cc:93
↓ 4 callersMethodResetAncestorId
tao/tao_bridge/passes/functionalize_cond.cc:218
↓ 4 callersMethodRun
tao/tao_bridge/passes/tao_mark_for_compilation_pass.cc:1766
↓ 4 callersFunctionSanitizeFileName
tao/tao_bridge/tf/util.cc:258
↓ 4 callersMethodSize
Returns the number of elements in a cluster.
tao/tao_bridge/tf/union_find.h:32
↓ 4 callersMethodSkip
Skip the specified number of samples of 128-bits in the current stream.
tao_compiler/mlir/xla/ral/context/custom_library/philox_random.h:137
↓ 4 callersMethodTensors
tao_compiler/mlir/disc/tools/disc-replay/record.h:35
↓ 4 callersFunctionWrapperIfEmpty
tao_compiler/mlir/xla/ral/context/tensorflow/tf_context_impl.cc:161
↓ 4 callersMethod__init__
(self)
tools/torch_quant/tests/models.py:35
↓ 4 callersMethod__init__
( self, c_in, c_hidden, no_heads, pair_bias=False, c_z=None,
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/msa.py:39
↓ 4 callersMethod_benchmark
( self, func_name, wrapper, traced_model, inputs, small_inputs
pytorch_blade/benchmark/detectron2/run_blade.py:261
↓ 4 callersFunction_default_device
()
examples/PyTorch/Inference/hf_transformers/blade_adapter.py:195
↓ 4 callersFunction_get_current_onnx_opset_version
()
pytorch_blade/torch_blade/pass_manager.py:27
↓ 4 callersFunction_jit_pass_add_placeholder_for_fake_quant
(c_module)
pytorch_blade/torch_blade/quantization/__init__.py:26
↓ 4 callersFunction_kwargs_to_args
(input_order, **kwargs)
examples/PyTorch/Inference/hf_transformers/blade_adapter.py:191
↓ 4 callersFunction_optimize_common
(c_module)
pytorch_blade/torch_blade/pass_manager.py:349
↓ 4 callersFunction_script_module_preprocess
(s_module, inputs, input_dims=[])
pytorch_blade/torch_blade/exporter.py:40
↓ 4 callersMethod_test_broadcast
(self, broadcast_func)
pytorch_blade/tests/disc/ops/test_broadcast.py:24
↓ 4 callersMethodadd_module
(self, full_path: str, module: nn.Module)
tools/torch_quant/torch_quant/graph.py:167
↓ 4 callersFunctionallElementsAreSameValue
Returns true if a tensor is constant and all its elements are equal to `elemVal`
tao_compiler/mlir/disc/transforms/disc_algebraic_simplifier.cc:38
↓ 4 callersFunctionallEquals
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:221
↓ 4 callersMethodalloc_persistent
tao_compiler/mlir/xla/ral/device/gpu/gpu_driver.cc:117
↓ 4 callersFunctionassert_almost_equal
(x, y, rtol=None, atol=None)
pytorch_blade/torch_blade/testing/common_utils.py:158
↓ 4 callersMethodbackward
(ctx: "Gather", grad_output: Tensor)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/distributed/comm.py:132
↓ 4 callersFunctionbenchmark
(backend, precision, model, inp, batch_size)
examples/PyTorch/Inference/CUDA/ResNet/main.py:71
↓ 4 callersFunctionbitonicTop128
tao_compiler/mlir/xla/ral/context/custom_library/top128.cu.h:20
↓ 4 callersFunctionbroadcastAs
tao_compiler/mlir/disc/transforms/mhlo_decomp_rewriters.cc:36
↓ 4 callersFunctionbuildFoldProducerExtractSlice
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:163
↓ 4 callersFunctionbuildLowerMultiLevelPackToLoop
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:202
↓ 4 callersFunctionbuildTileToForEachThreadOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:113
↓ 4 callersFunctionbuildVectorize
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:207
↓ 4 callersMethodc10Type
pytorch_blade/pytorch_blade/ltc/disc_backend/backend_impl.cpp:60
↓ 4 callersFunctioncalculate_md5
tao_compiler/mlir/xla/ral/context/common_context_impl_acl.cc:124
↓ 4 callersFunctioncompareSymbolicDimOpNames
Gives a consistent order of a list op SymbolicDim Ops
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:45
↓ 4 callersMethodcompute
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/operators/matmul.hpp:54
↓ 4 callersFunctioncreateDiscCanonicalizerPass
Create a Canonicalizer pass.
tao_compiler/mlir/disc/transforms/disc_canonicalizer.cc:58
↓ 4 callersFunctioncreateMemRef1DReinterpretCast
reinterpret_cast the input memref into 1D
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:1242
↓ 4 callersFunctioncreateSharedMemory
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:974
↓ 4 callersFunctioncreateSharedMemoryForOp
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:2335
↓ 4 callersMethoddesc
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:24
↓ 4 callersMethoddumped_file
Get full path of dumped file".
tao/tao_bridge/test/tao_ut_common.py:92
↓ 4 callersFunctionemitPOD
tao_compiler/mlir/disc/transforms/disc_lower_to_library_call.cc:681
↓ 4 callersFunctionevaluate_torch
(model, inputs)
examples/PyTorch/Inference/CUDA/BERT/main.py:111
↓ 4 callersMethodfeed_from
Fill the tensor with a src tensor TODO(xpz): may replace is_deconv_weights with a enum for other purposes
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:808
↓ 4 callersFunctionfillNCHW
tao_compiler/mlir/disc/transforms/conv_rewriter.cc:128
↓ 4 callersMethodforward
(self, w: torch.Tensor)
pytorch_blade/tests/tensorrt/test_tensorrt.py:32
↓ 4 callersFunctionfusionTypeFromString
Convert to a fusion type from its string representation.
tao_compiler/mlir/disc/transforms/fusion_utils.cc:205
↓ 4 callersMethodgetDimValue
tao_compiler/mlir/disc/transforms/shape_utils.cc:683
↓ 4 callersFunctiongetFusionFullName
Returns the full name of the fusion op Here full name is composed of the name and tag of the fusion op.
tao_compiler/mlir/disc/transforms/fusion_utils.cc:298
↓ 4 callersMethodgetFusionTypeStr
Returns the fusion kind of the fusion pattern.
tao_compiler/mlir/disc/transforms/fusion_utils.h:324
↓ 4 callersMethodgetHostConstant
tao_compiler/mlir/xla/ral/ral_metadata.cc:93
↓ 4 callersFunctiongetMemRefValueSymbolicDimRefs
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:760
↓ 4 callersFunctiongetNumAvailableCores
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:761
↓ 4 callersMethodgetOutputPlacements
tao_compiler/mlir/disc/transforms/mhlo_placer.cc:573
↓ 4 callersFunctiongetPDLDependentDialects
Adds related depedent dialects (e.g. PDL dialect).
tao_compiler/mlir/disc/transforms/disc_pdl_utils.cc:459
↓ 4 callersFunctiongetRankedValueSymbolicDimRefs
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:744
↓ 4 callersFunctiongetShapeCalcOperandList
tao_compiler/mlir/disc/transforms/placement_utils.cc:117
↓ 4 callersMethodgetTransposeOps
tao_compiler/mlir/disc/transforms/disc_transpose_simplifier.cc:298
↓ 4 callersFunctiongetValueUsersInFusionLike
returns the users of the `memref`. The users should be in the same fusion like `op`.
tao_compiler/mlir/disc/transforms/fusion_utils.cc:129
↓ 4 callersFunctiongetVectorSliceOpOffset
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:767
↓ 4 callersFunctiongetVectorizeOrTileHint
tao_compiler/mlir/disc/transforms/codegen_utils.cc:49
↓ 4 callersFunctiongetWeightPrePackingCacheCapacity
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.cc:152
↓ 4 callersMethodget_attr
(self, full_path: str)
tools/torch_quant/torch_quant/graph.py:162
↓ 4 callersMethodget_index
tao_compiler/mlir/xla/ral/context/custom_library/tf_topk.cu.h:54
↓ 4 callersMethodget_or_create_packed_weight
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.h:102
↓ 4 callersFunctionget_shape_for_one_input
(c_module, input_setting, trt_unsupported)
pytorch_blade/torch_blade/tensorrt/dynamic_shapes.py:37
↓ 4 callersFunctionget_simple_node_by_name
(simple_graph: SimpleGraph, name: str)
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:640
↓ 4 callersFunctioniceil
tao_compiler/mlir/xla/ral/context/custom_library/topk_util.cu.h:25
↓ 4 callersFunctioninferTransposeAttr
tao_compiler/mlir/disc/transforms/conv_rewriter.cc:168
↓ 4 callersMethodinitFusionPattern
tao_compiler/mlir/disc/transforms/fusion_utils.cc:1192
↓ 4 callersMethodinput_nodes_index
Get indexes of input nodes for node of given index.
tensorflow_blade/tf_blade/util/simple_graph.py:148
↓ 4 callersMethodinput_tensors
Find tensors that come into graph segment as input.
tensorflow_blade/tf_blade/util/simple_graph.py:364
↓ 4 callersFunctionisConstantMemRef
Returns true if the underlying buffer of this memref is a const buffer.
tao_compiler/mlir/disc/disc_util.cc:316
↓ 4 callersFunctionisLargeConcatOp
tao_compiler/mlir/disc/transforms/fusion_utils.cc:1345
↓ 4 callersFunctionisMarkShapeCalcTargetOp
Check Op if it is a mhlo Op or tensor Op.
tao_compiler/mlir/disc/transforms/placement_utils.h:98
↓ 4 callersFunctionisMhloDialect
Check Op if it is a mhlo Op.
tao_compiler/mlir/disc/transforms/placement_utils.h:75
↓ 4 callersFunctionisMutableType
pytorch_blade/pytorch_blade/compiler/jit/torch/alias_analysis.cpp:76
↓ 4 callersMethodisSameNumElements
tao_compiler/mlir/disc/transforms/shape_utils.cc:1596
↓ 4 callersFunctionisSparseFusion
tao_compiler/mlir/disc/transforms/fusion_utils.cc:537
↓ 4 callersFunctionisSupportedDot
tao_compiler/mlir/disc/transforms/fusion_utils_transform_based.cc:21
↓ 4 callersMethodisTransformBasedFusion
Returns true if the fusion type is transform-based fusion.
tao_compiler/mlir/disc/transforms/fusion_utils.h:345
↓ 4 callersMethodis_constant
tao/tao_bridge/executable.h:58
↓ 4 callersMethodis_platform_alibaba
(self)
tao/tao_bridge/test/gpu/mlir/base.py:54
↓ 4 callersMethodis_valid
tao/tao_bridge/executable.h:160
↓ 4 callersFunctionlink_dirs
(dst_dir, src_dir)
scripts/python/common_setup.py:550
↓ 4 callersFunctionlink_dirs
(dst_dir, src_dir)
pytorch_blade/common_setup.py:550
↓ 4 callersMethodmakeSubTaskFromRange
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:696
↓ 4 callersFunctionmatch_submodules
Iterate over the submodules and return name list of those whose type is matched in type_list. Args: model (nn.Module): Model to be m
pytorch_blade/torch_blade/exporter.py:117
↓ 4 callersFunctionmaybeConvert
return a new memref type with provided memory space if the input type if a memref type otherwise return the original type.
tao_compiler/mlir/disc/transforms/disc_assign_memory_space.cc:51
↓ 4 callersMethodminusOneFromDim
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:664
↓ 4 callersFunctionmkldnn_build_dir
(root=None)
scripts/python/common_setup.py:252
↓ 4 callersFunctionmkldnn_build_dir
(root=None)
pytorch_blade/common_setup.py:252
← previousnext →501–600 of 6,088, ranked by callers