MCPcopy Create free account

hub / github.com/alibaba/BladeDISC / functions

Functions6,088 in github.com/alibaba/BladeDISC

↓ 2 callersFunctionHistogramMapToRepeatedOpAndCount
tao/tao_bridge/tf/xla_cluster_util.cc:341
↓ 2 callersFunctionHumanReadableNumOps
tao/tao_bridge/tf/util.cc:156
↓ 2 callersMethodInit
tao/tao_bridge/kernels/tao_compilation_info_collector.cc:203
↓ 2 callersFunctionInitBladeDiscEngine
pytorch_blade/pytorch_blade/compiler/mlir/runtime/disc_engine.cpp:102
↓ 2 callersFunctionInitTorchBladeEngine
pytorch_blade/pytorch_blade/compiler/backends/engine_class.cpp:257
↓ 2 callersFunctionInitializeTrtPlugins
pytorch_blade/pytorch_blade/compiler/tensorrt/bridge/tensorrt_common.cpp:59
↓ 2 callersFunctionInlineCallInGraph
tao/tao_bridge/tf/lower_if_op.cc:206
↓ 2 callersFunctionInlineCallInGraph
tao/tao_bridge/tf/lower_while_op.cc:456
↓ 2 callersMethodInsert
Inserts `value` into the ordered set. Returns true if the value was not present in the set before the insertion.
tao_compiler/mlir/disc/utils/cycle_detector.h:48
↓ 2 callersMethodIsActivateNode
tao_compiler/mlir/disc/utils/cycle_detector.cc:342
↓ 2 callersFunctionIsCandidateBroadcastOp
tao_compiler/mlir/disc/transforms/disc_specialize_fusion_with_speculation.cc:38
↓ 2 callersMethodIsCompilableFunctionalOp
Assume `call_def` is a functional control flow ops
tao/tao_bridge/tf/compilability_check_util.h:192
↓ 2 callersFunctionIsEmptyBuffer
tao_compiler/mlir/xla/ral/context/tensorflow/tf_context_impl.cc:194
↓ 2 callersFunctionIsEnableReplayToolkit
pytorch_blade/pytorch_blade/ltc/disc_compiler/replay.cpp:185
↓ 2 callersFunctionIsInvalidTensorArrayOps
tao/tao_bridge/tf/compilability_check_util.cc:107
↓ 2 callersFunctionIsNonTensorOrTypeAnalyzed
pytorch_blade/pytorch_blade/compiler/mlir/converters/mhlo_conversion.cpp:128
↓ 2 callersFunctionIsPermutation
tao/tao_bridge/tf/util.cc:81
↓ 2 callersMethodIsReachableNonConst
tao/tao_bridge/tf/graphcycles.cc:341
↓ 2 callersFunctionIsSmallBuffer
tao_compiler/mlir/disc/disc_util.cc:30
↓ 2 callersFunctionLaunchOpHasKernelForDevice
Returns true if there is kernel of `TaoLaunch` or `TaoMlirLaunch` for the target device
tao/tao_bridge/tf/xla_op_registry.cc:58
↓ 2 callersFunctionLogLevelStrToInt
Parse log level (int) from environment variable (char*)
tao_compiler/mlir/xla/ral/ral_logging.cc:45
↓ 2 callersFunctionLowerHLOToSharedLibrary
tao_compiler/mlir/disc/disc_compiler.cc:915
↓ 2 callersFunctionMakeError
Make a Status with a code, error message and payload, and also send it to LOG(<log_severity>) using the given filename and line (unless should_log is
tao/tao_bridge/errors.cc:60
↓ 2 callersMethodMakeNotPredicate
tao/tao_bridge/tf/deadness_analysis.cc:383
↓ 2 callersMethodMakeOrPredicate
tao/tao_bridge/tf/deadness_analysis.cc:379
↓ 2 callersMethodMarkDead
tao/tao_bridge/passes/functionalize_cond.cc:225
↓ 2 callersFunctionMarkForCompilation
tao/tao_bridge/passes/tao_mark_for_compilation_pass.cc:2370
↓ 2 callersFunctionMayCallFunction
tao/tao_bridge/tf/xla_cluster_util.cc:316
↓ 2 callersMethodMerge
tao/tao_bridge/passes/tao_mark_for_compilation_pass.cc:570
↓ 2 callersFunctionMoveToList
tao/tao_bridge/tf/graphcycles.cc:282
↓ 2 callersFunctionMoveToList
Collects ranks of nodes in vector `src` to vector `dst`
tao_compiler/mlir/disc/utils/cycle_detector.cc:235
↓ 2 callersMethodName
tao/tao_bridge/tf/xla_op_registry.cc:505
↓ 2 callersMethodNewExecutable
tao/tao_bridge/executable.cc:339
↓ 2 callersFunctionNodeToString
tao/tao_bridge/tf/resource_operation_safety_analysis.cc:252
↓ 2 callersFunctionParseArgvFromString
Given a string containing flags, parse them into the XLA command line flags. The parse is best effort, and gives up on the first syntax error.
tao/tao_bridge/tf/parse_flags_from_env.cc:95
↓ 2 callersFunctionParseDataType
tao_compiler/mlir/disc/tests/mlir_test.cc:78
↓ 2 callersFunctionParseFlagsFromEnv
Call Flags::Parse(argc, argv, flag_list) against any as yet unrecognized flags passed in from the environment.
tao/tao_bridge/tf/parse_flags_from_env.cc:178
↓ 2 callersFunctionParseInteger
tao_compiler/mlir/xla/ral/ral_logging.cc:33
↓ 2 callersFunctionPickDeviceForXlaImpl
tao/tao_bridge/tf/device_util.cc:101
↓ 2 callersMethodPopulate
We populate the nodes along a special topological order where nodes having the same root frame are placed adjacent to each other. This grouping enabl
tao/tao_bridge/tf/deadness_analysis.cc:1373
↓ 2 callersMethodPredecessors
tao/tao_bridge/tf/graphcycles.cc:408
↓ 2 callersFunctionPrepareCompilerInput
tao/tao_bridge/kernels/tao_compilation_cache.cc:776
↓ 2 callersFunctionPrepareMLIRClustering
tao/tao_bridge/passes/tao_build_tao_op_pass.cc:151
↓ 2 callersFunctionReadDoubleFromEnvVar
pytorch_blade/pytorch_blade/common_utils/utils.cpp:105
↓ 2 callersFunctionReadFileBytes
ReadFileBytes reads a file as bytes
pytorch_blade/pytorch_blade/ltc/disc_compiler/passes/io.h:21
↓ 2 callersFunctionReadStringFromEnvVar
pytorch_blade/pytorch_blade/torch-mlir/lib/utils/env.cpp:47
↓ 2 callersMethodRecordComputationFinish
tao/tao_bridge/kernels/profiling.cc:41
↓ 2 callersFunctionRecursivelySetDevice
Clean all device information for nodes may have connection to `initial_nodes` through resource edge or colocate. This ensures Place can run successful
tao/tao_bridge/passes/tao_mark_for_compilation_pass.cc:2144
↓ 2 callersMethodReserve
tao/tao_bridge/tf/ordered_set.h:63
↓ 2 callersMethodReserve
tao_compiler/mlir/disc/utils/cycle_detector.h:71
↓ 2 callersFunctionRocmCurrentArch
tao_compiler/mlir/disc/transforms/disc_gpu_kernel_to_blob.cc:65
↓ 2 callersMethodRun
tao_compiler/mlir/disc/tests/mlir_test.cc:239
↓ 2 callersFunctionScalarTypeFromString
pytorch_blade/pytorch_blade/compiler/jit/shape_type_spec.cpp:43
↓ 2 callersFunctionScalarTypeToString
pytorch_blade/pytorch_blade/compiler/jit/shape_type_spec.cpp:30
↓ 2 callersMethodSetNodeData
tao/tao_bridge/tf/graphcycles.cc:145
↓ 2 callersMethodShutdown
tao_compiler/decoupling/tao_compiler_trace.cc:28
↓ 2 callersFunctionSort
tao/tao_bridge/tf/graphcycles.cc:270
↓ 2 callersFunctionSort
Sorts nodes in the vector according to their ranks. Small rank first.
tao_compiler/mlir/disc/utils/cycle_detector.cc:222
↓ 2 callersMethodStartTimeUS
tao/tao_bridge/kernels/process.cc:36
↓ 2 callersMethodStop
tao/tao_bridge/kernels/profiling.cc:49
↓ 2 callersFunctionTEST
static shape test case
tao_compiler/mlir/disc/tests/tensorflow_ops/const.cc:25
↓ 2 callersFunctionTensorIndexToFlat
tao_compiler/mlir/xla/ral/context/custom_library/tf_transpose.cu.h:52
↓ 2 callersMethodToString
pytorch_blade/pytorch_blade/ltc/include/torch/csrc/lazy/ts_backend/ops/to_copy.h:59
↓ 2 callersMethodTrtDynamicRanges
tensorflow_blade/src/tensorrt/bridge/tensorrt_onnx_parser.h:32
↓ 2 callersMethodUnionWith
tao/tao_bridge/tf/device_util.cc:37
↓ 2 callersFunctionWriteTextProtoToUniqueFile
tao/tao_bridge/tf/dump_graph.cc:63
↓ 2 callersMethod__init__
(self)
scripts/python/common_setup.py:28
↓ 2 callersMethod__init__
(self, c_m: int, c_z: int, c_hidden_msa_att: int, c_hidden_opm: int, c
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/evoformer.py:293
↓ 2 callersMethod__init__
( self, c_t, c_hidden_tri_att, c_hidden_tri_mul, no_blocks, no
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/template.py:410
↓ 2 callersMethod__init__
(self)
pytorch_blade/common_setup.py:28
↓ 2 callersMethod__init__
(self)
pytorch_blade/tests/quantization/__init__.py:28
↓ 2 callersMethod__init__
(self)
pytorch_blade/tests/tensorrt/test_holder_serialize.py:177
↓ 2 callersFunction_adapt_node_number_inputs
(graph, node)
pytorch_blade/torch_blade/clustering/support_group_conversion.py:124
↓ 2 callersFunction_adapt_node_number_outputs
(graph, node)
pytorch_blade/torch_blade/clustering/support_group_conversion.py:130
↓ 2 callersFunction_all_to_all
(tensor: Tensor, in_dim: int = -1, out_dim: int = -1)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/distributed/comm.py:78
↓ 2 callersFunction_as_tensor
(node_or_tensor: str)
tensorflow_blade/tf_blade/util/tf_util.py:323
↓ 2 callersMethod_check_launch_op
(self, fn, launch_op_name)
tao/tao_bridge/test/gpu/mlir/test_mlir.py:31
↓ 2 callersMethod_check_launch_op
(self, fn, launch_op_name)
tao/tao_bridge/test/gpu/mlir/test_mlir_transpose.py:31
↓ 2 callersMethod_check_trt_engine_output
(self, opt_model, input, type)
pytorch_blade/tests/tensorrt/test_tensorrt.py:38
↓ 2 callersMethod_clear
(self)
pytorch_blade/torch_blade/onnx_backends/backend_testbed.py:319
↓ 2 callersFunction_copy_list_container
(container_map, local, graph, node, val, container)
pytorch_blade/torch_blade/python_ir_analysis.py:33
↓ 2 callersFunction_create_graph_def_circle
-> Relu -- / \ # noqa: W605 px -> Add -> Mul Mul -> Relu /
tensorflow_blade/tests/util/simple_graph_test.py:76
↓ 2 callersFunction_deepcopy
(model)
pytorch_blade/torch_blade/exporter.py:112
↓ 2 callersFunction_disc_compile
Compiles the :attr:`fx_g` with Torchscript compiler. .. warning:: This API is experimental and likely to change. Args:
pytorch_blade/torch_blade/dynamo/__init__.py:25
↓ 2 callersFunction_dump_to_tempfile
(tmp_dir, dump_bytes)
pytorch_blade/torch_blade/mlir/disc_engine_conversion.py:28
↓ 2 callersFunction_flat_idx_to_idx
( flat_idx: int, dims: Tuple[int], )
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/utils.py:114
↓ 2 callersFunction_fuse_supported_subgraph
(graph, nodes_to_fuse)
pytorch_blade/torch_blade/clustering/support_fusion_group.py:33
↓ 2 callersFunction_gather
(tensor: Tensor, dim: int = -1)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/distributed/comm.py:38
↓ 2 callersFunction_get_disc_decomp
()
pytorch_blade/torch_blade/dynamo/__init__.py:108
↓ 2 callersFunction_get_nodes
(blocks)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/distributed/convert.py:122
↓ 2 callersFunction_get_submodules
(model)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/distributed/convert.py:109
↓ 2 callersFunction_get_target_key
(graph)
pytorch_blade/torch_blade/tools/low_precision_analysis.py:58
↓ 2 callersFunction_get_unsupported_nodes
(graph)
pytorch_blade/torch_blade/neural_engine/neural_engine_optimization.py:28
↓ 2 callersFunction_inputs_to_device
(inputs: Union[torch.Tensor, Callable, Tuple], device: torch.device)
examples/PyTorch/Inference/hf_transformers/blade_adapter.py:199
↓ 2 callersFunction_is_number
(val)
pytorch_blade/torch_blade/clustering/support_group_conversion.py:120
↓ 2 callersFunction_is_primitive
(val)
pytorch_blade/torch_blade/python_ir_analysis.py:23
↓ 2 callersFunction_is_tensor_producer
(node)
pytorch_blade/torch_blade/clustering/support_fusion_algorithm.py:268
↓ 2 callersMethod_iter_but_no_len
(o)
pytorch_blade/torch_blade/testing/common_utils.py:83
↓ 2 callersFunction_jit_add_fake_quant_for_weight
(c_module)
pytorch_blade/torch_blade/quantization/__init__.py:38
↓ 2 callersFunction_jit_pass_remove_all_placeholder
(c_module)
pytorch_blade/torch_blade/quantization/__init__.py:30
← previousnext →901–1,000 of 6,088, ranked by callers