MCPcopy Create free account

hub / github.com/alibaba/BladeDISC / functions

Functions6,088 in github.com/alibaba/BladeDISC

⨍ Functions 6,088◇ Types & classes 1,279↳ Endpoints 6
↓ 1 callersMethod_remove_optimization_replaced_nodes
Remove Nodes Those are designed to be replaced by the Optimization Pass. While do notice that: sometimes we do node replacement with
tensorflow_blade/tf_blade/util/tf_hierarchy_pattern_match.py:241
↓ 1 callersFunction_reorder_raise_exception_node
(graph)
pytorch_blade/torch_blade/pass_manager.py:233
↓ 1 callersFunction_replace_inplace_name
(graph)
pytorch_blade/torch_blade/pass_manager.py:412
↓ 1 callersFunction_replace_pythonop
(graph, node, saved_module, name)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/distributed/convert.py:41
↓ 1 callersFunction_replace_saveop
(graph, pair_node, load_module, name)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/distributed/convert.py:69
↓ 1 callersMethod_run
(self, command)
pytorch_blade/setup.py:97
↓ 1 callersMethod_segment_graph
Generate graph segments. Parameters ---------- supported_list: Set[str] A set contains all supported *op type*.
tensorflow_blade/tf_blade/util/simple_graph.py:664
↓ 1 callersMethod_segment_to_subgraph
(self)
pytorch_blade/torch_blade/onnx_backends/backend_testbed.py:429
↓ 1 callersFunction_set_annotate_args
(c_module, annotations)
pytorch_blade/torch_blade/pass_manager.py:324
↓ 1 callersFunction_subgraph_to_bytes
(subgraph, group_name)
pytorch_blade/torch_blade/mlir/disc_engine_conversion.py:134
↓ 1 callersFunction_subgraph_to_bytes
(subgraph, group_name)
pytorch_blade/torch_blade/onnx_backends/backend_conversion.py:102
↓ 1 callersFunction_symlink_force
(target, link_name)
pytorch_blade/bazel_build.py:40
↓ 1 callersMethod_test_add_fake_quant_for_weight
(self, model, inp, target_quant_info, target_output, target_graph)
pytorch_blade/tests/quantization/test_graph_process.py:242
↓ 1 callersMethod_test_rhs_scalar_func
(self, torch_func)
pytorch_blade/tests/disc/ops/test_binary_ops.py:114
↓ 1 callersMethod_test_tensorrt_type_cast
(self, type)
pytorch_blade/tests/tensorrt/test_tensorrt.py:68
↓ 1 callersMethod_test_tensorrt_type_normal
(self, model, opt_model, type)
pytorch_blade/tests/tensorrt/test_tensorrt.py:52
↓ 1 callersMethod_test_tensorrt_type_overflow
(self, model, opt_model, type)
pytorch_blade/tests/tensorrt/test_tensorrt.py:60
↓ 1 callersMethod_transition
(self, m, mask)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/ops.py:223
↓ 1 callersFunction_try_cast_graph_integer_inputs_to_i32
(graph)
pytorch_blade/torch_blade/onnx_backends/backend_conversion.py:39
↓ 1 callersMethod_v2_feature_forward
(self, x, z)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/embedder.py:205
↓ 1 callersFunction_validate_dynamic_inputs
(val)
pytorch_blade/torch_blade/config.py:88
↓ 1 callersFunction_validate_extra_dynamic_ranges
(val)
pytorch_blade/torch_blade/config.py:80
↓ 1 callersMethod_validate_pattern_and_set_root
(self)
tensorflow_blade/tf_blade/util/tf_hierarchy_pattern_match.py:181
↓ 1 callersMethod_wrap_up
(self, o: torch.Tensor, q_x: torch.Tensor)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/ops.py:343
↓ 1 callersFunctionacl_root_dir
(root)
scripts/python/common_setup.py:261
↓ 1 callersFunctionacl_root_dir
(root)
pytorch_blade/common_setup.py:261
↓ 1 callersFunctionaddAllRegisteredCanonicalizationPatterns
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:865
↓ 1 callersFunctionaddCanonicalizationPatterns
tao_compiler/mlir/disc/transforms/disc_transpose_simplifier.cc:85
↓ 1 callersFunctionaddCanonicalizationPatterns
Adds canonicalization patterns to the list of patterns.
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:648
↓ 1 callersFunctionaddPredefinedPrototypes
Combines the `chunkBuffer` with some pre-defined helper function prototypes. The result is written to a newly allocated buffer which will be returned.
tao_compiler/mlir/disc/transforms/disc_pdl_utils.cc:120
↓ 1 callersFunctionadd_arguments_common
(parser)
pytorch_blade/common_setup.py:684
↓ 1 callersFunctionadd_arguments_platform_alibaba
(parser)
scripts/python/common_setup.py:571
↓ 1 callersFunctionadd_arguments_platform_alibaba
(parser)
pytorch_blade/common_setup.py:571
↓ 1 callersFunctionadd_identity_n
( graph_def: tf.GraphDef, input_name_list: List[str], name: str, data_type_list: List[tf.DType
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:305
↓ 1 callersMethodadd_low_precision_module
(self, module_type)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/foldacc.py:64
↓ 1 callersFunctionadd_low_precision_observer
(optimized_c_module, mode)
pytorch_blade/torch_blade/tools/low_precision_analysis.py:33
↓ 1 callersFunctionadd_merge
( graph_def: tf.GraphDef, input_name_list: List[str], name: str, data_type: tf.DType )
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:338
↓ 1 callersFunctionadd_placeholder
( graph_def: tf.GraphDef, data_type: tf.DType, shape: List[Any], name: str )
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:413
↓ 1 callersFunctionadd_ral_link_if_not_exist
()
pytorch_blade/common_setup.py:669
↓ 1 callersFunctionadd_switch
( graph_def: tf.GraphDef, input_name_list: List[str], name: str, data_type: tf.DType )
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:530
↓ 1 callersMethodaddmm
(M, mat1, mat2)
pytorch_blade/tests/disc/ops/test_matmul.py:118
↓ 1 callersFunctionalignPtr
ALIGNPTR
tao_compiler/mlir/xla/ral/context/custom_library/topk_sort.cu.h:29
↓ 1 callersFunctionaligned_malloc
tao_compiler/mlir/xla/ral/context/common_context_impl.h:148
↓ 1 callersFunctionallClose
pytorch_blade/pytorch_blade/compiler/backends/engine_class.cpp:27
↓ 1 callersMethodallocate
tensorflow_blade/src/tensorrt/bridge/tensorrt_tf_allocator.cc:32
↓ 1 callersMethodamp
( self, model: nn.Module, act_ob_ctr: Optional[Callable[..., Observer]] = None,
tools/torch_quant/torch_quant/quantizer.py:156
↓ 1 callersMethodamp_gm
( self, name: str, gm: GraphModule, root: nn.Module, ob_types: ObserverTypes )
tools/torch_quant/torch_quant/quantizer.py:143
↓ 1 callersFunctionanalysis_python_ir
(outer_block)
pytorch_blade/torch_blade/python_ir_analysis.py:268
↓ 1 callersFunctionanalyze_target
(target, diff_percent, compare_fields)
pytorch_blade/benchmark/TorchBench/results_analysis.py:45
↓ 1 callersFunctionany_le
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/utils.hpp:82
↓ 1 callersFunctionapplyShapeComputationOptimization
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:1740
↓ 1 callersMethodarg_iterator
Gets an iterator to the arguments in the array.
tao_compiler/mlir/xla/ral/context/tensorflow/tf_context_impl.cc:368
↓ 1 callersFunctionarray_cmp
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/utils.hpp:284
↓ 1 callersFunctionassign
(d1, d2)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/utils.py:378
↓ 1 callersMethodassignSchedule
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:356
↓ 1 callersFunctionassumeAlignmentOnGPU
Assume the alignment of memrefs on GPU used in the given op.
tao_compiler/mlir/disc/transforms/disc_for_loop_unroll_interleave.cc:54
↓ 1 callersFunctionauto_detect_host_cpu
(args)
scripts/python/common_setup.py:361
↓ 1 callersFunctionauto_detect_host_cpu
(args)
pytorch_blade/common_setup.py:361
↓ 1 callersFunctionautomatic_chunk_size
(seq_len)
examples/PyTorch/Inference/CUDA/AlphaFold/unifold_inference.py:34
↓ 1 callersFunctionbackwardToFindValidValueToToReuse
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:2164
↓ 1 callersFunctionbatchTop128Launch
tao_compiler/mlir/xla/ral/context/custom_library/small_top128.cu.h:156
↓ 1 callersFunctionbatchedFilterNCompressGpu
tao_compiler/mlir/xla/ral/context/custom_library/filter_n_compress.cu.h:76
↓ 1 callersFunctionbatchedFilteredTop128Gpu
tao_compiler/mlir/xla/ral/context/custom_library/top128.cu.h:232
↓ 1 callersFunctionbatchedKthBiggestGpu
tao_compiler/mlir/xla/ral/context/custom_library/kth_biggest.cu.h:58
↓ 1 callersFunctionbatchedMultiSetsTop2Gpu
tao_compiler/mlir/xla/ral/context/custom_library/top2.cu.h:86
↓ 1 callersFunctionbatchedTop128Gpu
tao_compiler/mlir/xla/ral/context/custom_library/top128.cu.h:103
↓ 1 callersFunctionbenchmark
(model, device, n_warmup=10, n_repeat=20)
tools/torch_quant/bert_ptq_demo.py:85
↓ 1 callersFunctionbenchmark_ort
(model, inputs, backend, batch_size)
examples/PyTorch/Inference/CPU/albert/main.py:104
↓ 1 callersFunctionbetterToUseAlloca
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:72
↓ 1 callersFunctionbitonicExchange64_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:311
↓ 1 callersFunctionbitonicExchange64_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:302
↓ 1 callersFunctionbitonicMerge128_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:235
↓ 1 callersFunctionbitonicMerge128_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:226
↓ 1 callersFunctionbitonicMerge256_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:253
↓ 1 callersFunctionbitonicMerge256_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:244
↓ 1 callersFunctionbitonicMerge64_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:193
↓ 1 callersFunctionbitonicMerge64_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:184
↓ 1 callersFunctionbitonicSort128_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:332
↓ 1 callersFunctionbitonicSort128_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:319
↓ 1 callersFunctionbitonicSort32_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:132
↓ 1 callersFunctionbitonicSort32_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:112
↓ 1 callersFunctionbitonicSort64_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:214
↓ 1 callersFunctionbitonicSort64_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:202
↓ 1 callersFunctionblade_trt_optimize
(model, inputs, fp16: bool, is_static: bool, out_file: str)
examples/PyTorch/Inference/CUDA/BERT/main.py:230
↓ 1 callersFunctionbufferizeDestinationStyleOpInterface
Generic conversion for any DestinationStyleOpInterface on tensors.
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:888
↓ 1 callersFunctionbufferizeTensorEmptyOps
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:149
↓ 1 callersFunctionbuildConvertPaddingPlaceholderToConstOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:278
↓ 1 callersMethodbuildEqualShapesInFusion
tao_compiler/mlir/disc/transforms/shape_utils.cc:1797
↓ 1 callersMethodbuildGuardCondition
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:849
↓ 1 callersFunctionbuildInlineReductionInitializerOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:249
↓ 1 callersFunctionbuildLinalgEagerlyBackwardInitTensorOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:286
↓ 1 callersFunctionbuildLowerConditionalGenericOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:332
↓ 1 callersFunctionbuildMin
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:1325
↓ 1 callersMethodbuildProducerGraph
Returns a producer graph for ops in the block: map<op-in-the-block, direct-or-indirect-producers-set-of-this-op>
tao_compiler/mlir/disc/transforms/disc_transpose_simplifier.cc:326
↓ 1 callersFunctionbuildQuantizedDotGeneralOp
tao_compiler/mlir/disc/transforms/disc_convert_fake_quant_op.cc:160
↓ 1 callersFunctionbuildReductionInputFuseOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:308
↓ 1 callersFunctionbuildReductionOutputFuseOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:301
↓ 1 callersFunctionbuildReplaceConstPaddingValueOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:271
↓ 1 callersMethodbuildTieShapeOps
After we do all shape-related simplifications in the tensor level, we build explicit connection between all symbolic dims using 'disc_shape.tie_shape'
tao_compiler/mlir/disc/transforms/shape_utils.cc:1531
↓ 1 callersFunctionbuildVectorizeConditionalGenericOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:317
← previousnext →1,801–1,900 of 6,088, ranked by callers