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
↓ 4 callersMethodname2index
Get index of node.
tensorflow_blade/tf_blade/util/simple_graph.py:118
↓ 4 callersMethodnew_sess
(self, allow_growth=True, allocator_type='BFC', log_device_placement=False,
tao/tao_bridge/test/tao_ut_common.py:114
↓ 4 callersFunctionnextWorkspacePtr
NEXTWORKSPACEPTR
tao_compiler/mlir/xla/ral/context/custom_library/topk_sort.cu.h:38
↓ 4 callersMethodoptimize_graph_def
( self, graph_def: tf.GraphDef, model_outputs: List[str], test_data: List[Any]
tensorflow_blade/tf_blade/gpu/tf_to_trt.py:260
↓ 4 callersMethodoutput_nodes
Get names of output nodes, which are those without downstream.
tensorflow_blade/tf_blade/util/simple_graph.py:144
↓ 4 callersFunctionoverrideCanFuseOnCPULegacy
pytorch_blade/pytorch_blade/ltc/disc_compiler/passes/graph_fuser.cpp:1285
↓ 4 callersFunctionparseConvParams
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.h:364
↓ 4 callersFunctionpopulateDiscPdlPatternsFromFiles
Adds customized pdl patterns
tao_compiler/mlir/disc/transforms/disc_pdl_utils.cc:485
↓ 4 callersFunctionpopulateDiscPdlPatternsFromString
Parses pdll patterns from string, compile them and then add to `patterns`.
tao_compiler/mlir/disc/transforms/disc_pdl_utils.cc:464
↓ 4 callersFunctionprecision_to_dtype
(pr)
pytorch_blade/benchmark/torch-tensorrt/perf_run.py:360
↓ 4 callersMethodprint
tao_compiler/mlir/disc/tools/disc-transform/LinalgExt/LinalgExtOps.cc:492
↓ 4 callersFunctionprintStats
(backend, timings, precision, batch_size=1)
pytorch_blade/benchmark/torch-tensorrt/perf_run.py:318
↓ 4 callersFunctionral_base_cpu_bitcast_update_ref_count
tao_compiler/mlir/xla/ral/context/base/cpu/cpu_context_impl.cc:234
↓ 4 callersFunctionral_base_cuda_bitcast_update_ref_count
tao_compiler/mlir/xla/ral/context/base/cuda/cuda_context_impl.cc:486
↓ 4 callersFunctionral_tf_bitcast_update_ref_count
tao_compiler/mlir/xla/ral/context/tensorflow/tf_context_impl.cc:653
↓ 4 callersMethodraw_dealloc
tao_compiler/mlir/xla/ral/device/gpu/gpu_driver.cc:145
↓ 4 callersMethodrelease
Release the ownership of the wrapper buffer. This requires that the buffer is owned by this wrapper.
tao_compiler/mlir/xla/ral/context/base/base_context.h:64
↓ 4 callersFunctionrunCanonicalizer
tao_compiler/mlir/disc/transforms/disc_shape_optimization.cc:654
↓ 4 callersFunctionrun_py_test
(py_bin, test_root, output_file, includes, envs=[])
scripts/python/tao_build.py:499
↓ 4 callersFunctionset_attr_bool
(node: tf.NodeDef, key: str, value: bool)
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:96
↓ 4 callersFunctionset_attr_string
(node: tf.NodeDef, key: str, value: str)
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:154
↓ 4 callersFunctionset_env
(**environ)
pytorch_blade/tests/disc/testing_base.py:129
↓ 4 callersMethodset_opts
tao/tao_bridge/passes/tao_defuse_pass.h:28
↓ 4 callersFunctionsplit
pytorch_blade/pytorch_blade/common_utils/utils.cpp:23
↓ 4 callersFunctiontao_bazel_dir
(root=None)
scripts/python/tao_build.py:100
↓ 4 callersMethodto_default_format
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:747
↓ 4 callersMethodto_graphdef
Generate subgraph GraphDef from this segment. Parameters ---------- seg_id : int = 0 The id of this segment/subgr
tensorflow_blade/tf_blade/util/simple_graph.py:403
↓ 4 callersMethodto_grouped
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:244
↓ 4 callersFunctiontry_get_const_or_enter_node_value
( graph_def: tf.GraphDef, simple_graph: SimpleGraph, name: str )
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:1014
↓ 4 callersMethodunion
(self, x: int, y: int)
pytorch_blade/torch_blade/algorithm/union_set.py:41
↓ 4 callersMethodupdateKnownNonNegative
tao_compiler/mlir/disc/IR/disc_shape_ops.cc:333
↓ 4 callersFunctionuseTransformSchedule
Returns true if `DISC_ENABLE_TRANSFORM_SCHEDULE` is true
tao_compiler/mlir/disc/disc_util.cc:222
↓ 4 callersFunctionwalkRankedTensorValue
Walk each ranked tensor type values inside op.
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:824
↓ 3 callersFunctionBuildRetvalNode
tao/tao_bridge/passes/functionalize_control_flow_util.cc:46
↓ 3 callersFunctionCheckClusterDependencyForCycles
Adds an edge from ancestor to successor to the cycle detector, and returns an error if that edge causes the formation of a cycle. In the error case, l
tao/tao_bridge/passes/tao_encapsulate_subgraphs_pass.cc:2200
↓ 3 callersFunctionCompileFunctionImpl
tao/tao_bridge/kernels/tao_compilation_cache.cc:921
↓ 3 callersFunctionConvertArrayAttrToInt
tao_compiler/mlir/disc/transforms/disc_pdl_utils.cc:53
↓ 3 callersFunctionCreateGraph
tao/tao_bridge/passes/tao_bace_reformat_pass_test.cc:28
↓ 3 callersFunctionCreateReplayRecord
tao_compiler/mlir/disc/tools/disc-replay/record.h:46
↓ 3 callersMethodDebugString
tao/tao_bridge/tf/device_util.cc:81
↓ 3 callersFunctionDiscFusion
pytorch_blade/pytorch_blade/ltc/disc_compiler/passes/disc_fuser.cpp:97
↓ 3 callersFunctionDumpFunctionDefToFile
tao/tao_bridge/tf/dump_graph.cc:106
↓ 3 callersFunctionDumpGraph
tao/tao_bridge/passes/functionalize_control_flow.cc:58
↓ 3 callersFunctionDumpIValues
pytorch_blade/pytorch_blade/common_utils/utils.cpp:63
↓ 3 callersMethodFindBranchOf
tao/tao_bridge/passes/functionalize_cond.cc:886
↓ 3 callersFunctionFindNodeByName
tao/tao_bridge/passes/tao_partially_decluster_pass_test.cc:118
↓ 3 callersMethodFindPath
tao/tao_bridge/tf/graphcycles.cc:297
↓ 3 callersFunctionGetBackendName
pytorch_blade/pytorch_blade/compiler/mlir/runtime/disc_engine.cpp:98
↓ 3 callersFunctionGetBatchSize
tao_compiler/mlir/xla/ral/context/stream_executor_based_impl.cc:532
↓ 3 callersFunctionGetGlobalJitLevelForGraph
tao/tao_bridge/tf/xla_cluster_util.cc:292
↓ 3 callersFunctionGetGpuDeviceUUID
tao/tao_bridge/cuda_utils.cc:24
↓ 3 callersMethodGetNodeData
tao/tao_bridge/tf/graphcycles.cc:141
↓ 3 callersFunctionGetNodeOrFuncAttr
Returns true iff the attribute `attr_name` is attached to either the node or to it's callee.
tao/tao_bridge/passes/tao_mark_for_compilation_pass.cc:1089
↓ 3 callersFunctionGetNodesRelatedToRefVariablesInDirection
Helper for GetNodesRelatedToRefVariables that traverses the graph in one direction.
tao/tao_bridge/tf/xla_cluster_util.cc:492
↓ 3 callersFunctionGetNumericLimits
tao_compiler/mlir/disc/transforms/lower_tf.cc:251
↓ 3 callersMethodGetSendFromHostNode
tao/tao_bridge/passes/tao_encapsulate_subgraphs_pass.cc:743
↓ 3 callersFunctionGetSwitchPredicate
Returns the predicate of a switch.
tao/tao_bridge/passes/functionalize_cond.cc:85
↓ 3 callersFunctionGetTransposeOutputType
tao_compiler/mlir/disc/transforms/disc_dense_to_sparse.cc:114
↓ 3 callersFunctionInitializeTrtPlugins
tensorflow_blade/src/tensorrt/bridge/tensorrt_common.cpp:58
↓ 3 callersMethodInsert
tao/tao_bridge/tf/device_util.cc:26
↓ 3 callersMethodIsEmpty
tao/tao_bridge/passes/functionalize_cond.cc:132
↓ 3 callersFunctionIsMemRefAliasOp
tao_compiler/mlir/disc/disc_util.cc:66
↓ 3 callersMethodIsReachable
tao/tao_bridge/tf/graphcycles.cc:337
↓ 3 callersFunctionIsShapeConsumerOp
tao/tao_bridge/tf/xla_cluster_util.cc:328
↓ 3 callersFunctionIsSmallCpuBuffer
tao_compiler/mlir/disc/disc_util.cc:44
↓ 3 callersFunctionLaunchOpHasKernelForDeviceImpl
tao/tao_bridge/tf/xla_op_registry.cc:41
↓ 3 callersFunctionMakeBaseCudaContext
tao_compiler/mlir/xla/ral/context/base/cuda/cuda_context_impl.cc:161
↓ 3 callersMethodNewName
tao/tao_bridge/tf/lower_if_op.cc:137
↓ 3 callersMethodOnEvent
tao_compiler/decoupling/tao_compiler_trace.cc:41
↓ 3 callersFunctionPartiallyDeclusterGraph
Clones nodes to outside their cluster to avoid device-to-host copies. For instance, converts this: ..... | v A_Clustered ====> C_Unclustered | v B_C
tao/tao_bridge/passes/tao_partially_decluster_pass.cc:192
↓ 3 callersMethodPlacements
tao_compiler/mlir/disc/tools/disc-replay/record.h:36
↓ 3 callersFunctionProduct
tao/tao_bridge/tf/util.cc:212
↓ 3 callersFunctionPropagateInputShapes
pytorch_blade/pytorch_blade/compiler/jit/torch/shape_analysis.cpp:2695
↓ 3 callersFunctionQuantVerify
tao_compiler/mlir/disc/IR/hlo_disc_ops.cc:121
↓ 3 callersFunctionReachableDefinitions
tao/tao_bridge/tao_util.cc:34
↓ 3 callersFunctionRemoveFromXlaCluster
tao/tao_bridge/tf/xla_cluster_util.cc:222
↓ 3 callersMethodRun
tao/tao_bridge/kernels/launch_base.h:58
↓ 3 callersMethodRun
tao_compiler/mlir/disc/tools/disc-replay/disc_interpreter.cc:83
↓ 3 callersFunctionRunPass
tao/tao_bridge/passes/tao_bace_reformat_pass_test.cc:66
↓ 3 callersMethodSerialize
pytorch_blade/pytorch_blade/compiler/jit/shape_type_spec.cpp:57
↓ 3 callersMethodStart
tao/tao_bridge/tf/subprocess.cc:127
↓ 3 callersMethodStart
tao/tao_bridge/kernels/profiling.cc:21
↓ 3 callersFunctionStreamExecutorConvLayoutsToMetadata
tao_compiler/mlir/xla/ral/context/stream_executor_based_impl.cc:913
↓ 3 callersFunctionToCpu
tao/tao_bridge/kernels/tao_compilation_cache.cc:539
↓ 3 callersFunctionTryMergeNode
tao_compiler/mlir/disc/disc_util.cc:166
↓ 3 callersFunctionUint32sToUint64
tao_compiler/mlir/xla/ral/context/custom_library/random.h:37
↓ 3 callersFunctionUint64ToUint32s
The following 2 functions use the contract "lower 32 bits for the first uint32_t, higher 32 bits for the second". Note that this is endian-neutral, un
tao_compiler/mlir/xla/ral/context/custom_library/random.h:31
↓ 3 callersFunctionVerify
tao_compiler/mlir/disc/IR/hlo_disc_ops.cc:35
↓ 3 callersMethodWriteBytesToFile
pytorch_blade/pytorch_blade/common_utils/tempfs.cpp:79
↓ 3 callersFunction_add_module
(root: nn.Module, full_path: str, module: nn.Module)
tools/torch_quant/torch_quant/graph.py:67
↓ 3 callersMethod_appendNode
(self, old_node)
pytorch_blade/torch_blade/onnx_backends/backend_testbed.py:418
↓ 3 callersMethod_calculate_qparams
(self, min_val, max_val)
tools/torch_quant/torch_quant/observer.py:133
↓ 3 callersFunction_check_list_in_node
(node)
pytorch_blade/torch_blade/tools/onnx_lower_guard.py:16
↓ 3 callersMethod_conv2d
( self, x: tf.Tensor, w: tf.Tensor, b: tf.Tensor, strides: int = 1 )
tensorflow_blade/tests/tf_test_common.py:22
↓ 3 callersFunction_conv_type
(name: str, base: Type)
tools/torch_quant/torch_quant/observed_module.py:30
↓ 3 callersMethod_create_node
(self)
tensorflow_blade/tests/util/simple_graph_test.py:32
↓ 3 callersFunction_deduce_from_version_file
(cuda_home)
scripts/python/common_setup.py:432
↓ 3 callersFunction_deduce_from_version_file
(cuda_home)
pytorch_blade/common_setup.py:432
↓ 3 callersMethod_do_int8_calibration_optimzie
(self, model, calib_data)
pytorch_blade/tests/tensorrt/test_optimize.py:256
↓ 3 callersMethod_eval_models
(self, blade_model: WrapperModel, eager_model: PreTrainedModel)
examples/PyTorch/Inference/hf_transformers/tests/test_optimize.py:27
← previousnext →601–700 of 6,088, ranked by callers