MCPcopy Create free account

hub / github.com/alibaba/BladeDISC / functions

Functions6,088 in github.com/alibaba/BladeDISC

↓ 2 callersFunction_jit_replace_aten_fake_quant_with_custom_version
(c_module)
pytorch_blade/torch_blade/quantization/__init__.py:34
↓ 2 callersMethod_load_model
(self, type)
pytorch_blade/tests/tensorrt/test_tensorrt.py:24
↓ 2 callersMethod_locate_tao_compiler
Try to find tao_compiler binary under bazel build path if not specified by user or found already.
tao/tao_bridge/test/tao_ut_common.py:32
↓ 2 callersFunction_opt
(opt, value, cmd="build")
scripts/python/tao_build.py:181
↓ 2 callersFunction_optimize
(model, allow_tracing, model_inputs)
pytorch_blade/torch_blade/optimization.py:35
↓ 2 callersFunction_register_buffer
(root: nn.Module, full_path: str, tensor: torch.Tensor)
tools/torch_quant/torch_quant/graph.py:72
↓ 2 callersMethod_relpos_indices
( self, res_id: torch.Tensor, sym_id: Optional[torch.Tensor] = None, asym_id:
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/embedder.py:64
↓ 2 callersMethod_run_fake_quant
(self, axis)
tao/tao_bridge/test/gpu/mlir/test_mlir_fake_quant.py:31
↓ 2 callersMethod_setup_dump_graph
(testcase='')
tao/tao_bridge/test/tao_ut_common.py:73
↓ 2 callersMethod_setup_tf_logging
()
tao/tao_bridge/test/tao_ut_common.py:66
↓ 2 callersFunction_split
(tensor: Tensor, dim: int = -1)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/distributed/comm.py:27
↓ 2 callersFunction_submodule_names
If module name is 'foo', turn full path 'foo.bar.name' into 'bar.name'
tools/torch_quant/torch_quant/module.py:72
↓ 2 callersMethod_test_context
(self, flag_context_func, get_flag_func)
pytorch_blade/tests/test_flags.py:28
↓ 2 callersMethod_test_fake_quant_params
(self, fake_quant_node, target_val)
pytorch_blade/tests/quantization/__init__.py:95
↓ 2 callersMethod_test_get_set
(self, get_flag_func, set_flag_func, new_flag)
pytorch_blade/tests/test_flags.py:20
↓ 2 callersMethod_test_group_norm
(self, groupnorm)
pytorch_blade/tests/disc/ops/test_group_norm.py:19
↓ 2 callersMethod_test_linear
(self, dtype)
pytorch_blade/tests/disc/ops/test_matmul.py:41
↓ 2 callersMethod_test_logic_func
(self, torch_func)
pytorch_blade/tests/disc/ops/test_binary_ops.py:192
↓ 2 callersMethod_test_numeric_config
(self, attr, val, error_val)
pytorch_blade/tests/test_config.py:55
↓ 2 callersFunction_update_module
(modules, target, new_module)
tools/torch_quant/torch_quant/module.py:166
↓ 2 callersFunction_validate_dynamic_ranges
(val)
pytorch_blade/torch_blade/config.py:72
↓ 2 callersMethodaccept
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:1427
↓ 2 callersFunctionaddBoolFlags
tao_compiler/mlir/disc/tests/mlir_feature_test.cc:231
↓ 2 callersMethodaddDriver
tao_compiler/mlir/xla/ral/ral_context.cc:193
↓ 2 callersFunctionaddTransformDialectDependentDialects
tao_compiler/mlir/disc/tools/disc-transform/transforms/transform_dialect_interpreter.cc:54
↓ 2 callersFunctionadd_arguments_common
(parser)
scripts/python/common_setup.py:684
↓ 2 callersMethodadd_edge
(self, src: int, dst: int)
pytorch_blade/torch_blade/algorithm/directed_graph.py:75
↓ 2 callersMethodadd_edge
(self, src: int, dst: int)
pytorch_blade/torch_blade/clustering/support_fusion_algorithm.py:34
↓ 2 callersFunctionadd_identity
( graph_def: tf.GraphDef, input_name: str, name: str, data_type: tf.DType )
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:297
↓ 2 callersFunctionallConstIndexValues
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:225
↓ 2 callersFunctionallIndicesZeros
tao_compiler/mlir/disc/transforms/fusion_utils.cc:2691
↓ 2 callersFunctionapplyTypes
pytorch_blade/pytorch_blade/compiler/jit/torch/shape_analysis.cpp:150
↓ 2 callersMethodassert_pipeline_outputs_equal
(self, a, b)
examples/PyTorch/Inference/hf_transformers/tests/test_pipeline.py:53
↓ 2 callersFunctionattachSymbolicDimOpRefArrayAttrOnOperation
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:790
↓ 2 callersFunctionbackwardBcastPermutation
tao_compiler/mlir/disc/transforms/disc_transpose_simplifier.cc:340
↓ 2 callersFunctionbazel_test
(target, flag="")
scripts/python/tao_build.py:389
↓ 2 callersFunctionbenchmark
(model, inp, backend, batch_size)
examples/PyTorch/Inference/CPU/albert/main.py:131
↓ 2 callersFunctionbenchmark
(func_name, backend, enable_fp16, model, inp, batch_size)
pytorch_blade/benchmark/detectron2/run_blade.py:73
↓ 2 callersMethodbindValueNames
tao_compiler/mlir/disc/utils/source_emitter.cc:624
↓ 2 callersFunctionbitonicExchange32_256
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:294
↓ 2 callersFunctionbitonicExchange32_256_pair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:285
↓ 2 callersMethodbmm
(x, y)
pytorch_blade/tests/disc/ops/test_matmul.py:109
↓ 2 callersFunctionbuildFuseOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:127
↓ 2 callersFunctionbuildLinalgFuseProducersOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:264
↓ 2 callersFunctionbuildSymbolDimInstancesDominantMap
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:1462
↓ 2 callersFunctionbuildValueMapping
tao_compiler/mlir/disc/transforms/disc_transform_legalize_to_loop.cc:79
↓ 2 callersFunctionbuild_tao_compiler_add_flags_platform_alibaba
(root, args, flag)
scripts/python/common_setup.py:594
↓ 2 callersMethodcalculate_qparams
(self)
tools/torch_quant/torch_quant/observer.py:696
↓ 2 callersFunctioncalibrate
(model, dataloader, device)
tools/torch_quant/bert_ptq_demo.py:46
↓ 2 callersFunctioncheckMajorDimsAllOnes
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:782
↓ 2 callersFunctioncheck_tensor
(outputs, target, tolerance=1e-3)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/utils.py:55
↓ 2 callersFunctioncheck_two_shape
shape1: [shapes_for_subgraph1, shapes_for_subgraph2, ...] shape2: [shapes_for_subgraph1, shapes_for_subgraph2, ...] three conditions must
pytorch_blade/torch_blade/tensorrt/dynamic_shapes.py:96
↓ 2 callersFunctionclamp
tao_compiler/mlir/disc/transforms/disc_dynamic_slice_converter.cc:40
↓ 2 callersFunctioncloneAndMoveTo
tao_compiler/mlir/disc/transforms/disc_outline_cpu_kernel.cc:247
↓ 2 callersFunctioncollect_engines
Collect all engines in a script_module of the group type
pytorch_blade/torch_blade/utils.py:170
↓ 2 callersFunctioncollect_tensors
(data)
examples/PyTorch/Inference/CPU/albert/main.py:156
↓ 2 callersFunctioncompilePDLL
Compiles a pdll module into a pdl pattern module.
tao_compiler/mlir/disc/transforms/disc_pdl_utils.cc:137
↓ 2 callersFunctioncomputeShapeRatio
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:1553
↓ 2 callersFunctionconfig_mkldnn
(root, args)
scripts/python/common_setup.py:269
↓ 2 callersFunctionconvertGenericOpToConditionalGenericOp
tao_compiler/mlir/disc/tools/disc-transform/TransformOps/TransformOpsExt.cc:2469
↓ 2 callersFunctionconvertIfIntegerType
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:1374
↓ 2 callersFunctionconvert_save_model
(model)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/optimization/distributed/convert.py:130
↓ 2 callersFunctioncopyWithMemorySpace
Returns a new memref type with provided memory space
tao_compiler/mlir/disc/transforms/placement_utils.cc:256
↓ 2 callersMethodcpp_run
(self)
pytorch_blade/setup.py:90
↓ 2 callersFunctioncreateDiscAlgebraicSimplifierPass
tao_compiler/mlir/disc/transforms/disc_algebraic_simplifier.cc:596
↓ 2 callersFunctioncreateDiscLowerTfPass
Lower some tf ops before tf2mhlo lowering.
tao_compiler/mlir/disc/transforms/lower_tf.cc:1578
↓ 2 callersFunctioncreateDiscMemRefLoadStoreSimplifierPass
tao_compiler/mlir/disc/transforms/disc_memref_load_store_simplifier.cc:160
↓ 2 callersFunctioncreateDiscShapeSimplifierPass
tao_compiler/mlir/disc/transforms/disc_shape_simplifier.cc:537
↓ 2 callersFunctioncreateDiscSpecializeFusionWithSpeculationPass
tao_compiler/mlir/disc/transforms/disc_specialize_fusion_with_speculation.cc:750
↓ 2 callersFunctioncreateInitialValueForReduceOp
pytorch_blade/pytorch_blade/torch-mlir/lib/Conversion/TorchToMhlo/DiscTorchToMhlo.cpp:953
↓ 2 callersFunctioncreateLinalgCopyOp
Create a linalg::GenericOp version of an n-D copy that can further tile, lower to loops or vectorize, unlike the current implementation of memref::Cop
tao_compiler/mlir/disc/tools/disc-transform/utils.cc:51
↓ 2 callersFunctioncreate_engine
pytorch_blade/pytorch_blade/compiler/backends/engine_class.cpp:343
↓ 2 callersMethodcreate_ffg
(self)
pytorch_blade/tests/tensorrt/test_trt_support_fusion.py:24
↓ 2 callersFunctioncreate_method_from_graph
pytorch_blade/pytorch_blade/compiler/jit/tool_funcs.cpp:49
↓ 2 callersMethodcreate_wrapper_model
(opt_model: torch.jit.ScriptModule, info: ModelInfo)
examples/PyTorch/Inference/hf_transformers/blade_adapter.py:234
↓ 2 callersFunctiondeduce_cuda_info
Deduce cuda major and minor version and cuda directory.
scripts/python/common_setup.py:429
↓ 2 callersFunctiondelinearize
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:288
↓ 2 callersFunctiondiscEnableGlobalConstantStore
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:147
↓ 2 callersFunctiondisc_conf_file
(root)
scripts/python/common_setup.py:726
↓ 2 callersFunctiondisc_conf_file
(root)
pytorch_blade/common_setup.py:726
↓ 2 callersFunctiondivide
(numerator, denominator)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/distributed/comm.py:23
↓ 2 callersFunctiondtype
pytorch_blade/pytorch_blade/ltc/include/torch/csrc/lazy/ts_backend/ops/cast.h:35
↓ 2 callersMethodeffectiveSize
Returns the effective size (e.g. not counting const ops) of the ops this fusion pattern contains.
tao_compiler/mlir/disc/transforms/fusion_utils.cc:617
↓ 2 callersFunctionelementalLowerImplForBroadcastInDimOps
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:310
↓ 2 callersFunctionelementalLowerIota
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:999
↓ 2 callersMethodemitHeader
tao_compiler/mlir/xla/ral/ral_metadata.cc:126
↓ 2 callersFunctionemitInThreadReduction
emit in-thread reduction, and other non-dominant nodes including elementwise and column reduction
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:1004
↓ 2 callersFunctionemitPerElementReductionImpl
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:985
↓ 2 callersFunctionemitReduceMapOp
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:454
↓ 2 callersMethodemitTailer
tao_compiler/mlir/xla/ral/ral_metadata.cc:164
↓ 2 callersFunctionemitWarpReduce
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:3124
↓ 2 callersFunctionensureDenormalState
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:104
↓ 2 callersFunctionenvValueIsTrue
tao_compiler/mlir/disc/transforms/fusion_utils.cc:55
↓ 2 callersFunctionevaluate_torch
(model, inputs)
examples/PyTorch/Inference/CUDA/T5/main.py:47
↓ 2 callersFunctionevaluate_torch
(model, inputs)
examples/PyTorch/Inference/CUDA/S2T/main.py:65
↓ 2 callersFunctionexpandValueDims
Expand v in order to be broacast compatiable with the target ranked value.
tao_compiler/mlir/disc/transforms/lower_tf.cc:289
↓ 2 callersMethodextractContinuousDimEqualInfo
tao_compiler/mlir/disc/transforms/shape_utils.cc:1618
↓ 2 callersFunctionfillHWIO
tao_compiler/mlir/disc/transforms/conv_rewriter.cc:152
↓ 2 callersFunctionfillOHWI
tao_compiler/mlir/disc/transforms/conv_rewriter.cc:160
↓ 2 callersFunctionfillOIHW
tao_compiler/mlir/disc/transforms/conv_rewriter.cc:144
← previousnext →1,001–1,100 of 6,088, ranked by callers