MCPcopy Create free account

hub / github.com/PaddlePaddle/Paddle / functions

Functions118,755 in github.com/PaddlePaddle/Paddle

↓ 43 callersFunctionthread_id
paddle/phi/api/profiler/event.h:63
↓ 42 callersFunctionCUDAErrorCheck
paddle/fluid/framework/new_executor/instruction/instruction_util.h:72
↓ 42 callersMethodEnableTensorRtEngine
paddle/fluid/inference/api/analysis_config.cc:783
↓ 42 callersFunctionGetGPUDeviceCount
paddle/phi/backends/gpu/rocm/rocm_info.cc:79
↓ 42 callersMethodInt32Attr
paddle/fluid/pir/drr/src/pattern_context.cc:193
↓ 42 callersMethodSetNotOwned
paddle/fluid/framework/ir/graph.h:189
↓ 42 callersFunctionadd
paddle/phi/core/platform/monitor.h:95
↓ 42 callersMethodas_var_ref
paddle/cinn/ir/ir_base.cc:351
↓ 42 callersMethodcheck_grad
( self, inputs_to_check, output_names, no_grad_set=None, numeric_grad_
test/xpu/op_test_xpu.py:134
↓ 42 callersMethodcheck_operation
(self, a, b, c, op)
test/legacy_test/test_tensor_scalar_type_promotion_dynamic.py:33
↓ 42 callersMethodcheck_operation
(self, a, b, c, op)
test/legacy_test/test_tensor_scalar_type_promotion_static.py:37
↓ 42 callersFunctioncreate_test_act_stride_class
(base_class, api_name, paddle_api, numpy_api)
test/legacy_test/test_activation_stride_op.py:60
↓ 42 callersFunctiondemangle
paddle/common/enforce.h:46
↓ 42 callersMethodemplace_back
paddle/phi/core/memory/allocation/inlined_vector.h:40
↓ 42 callersMethodemplace_back
paddle/phi/core/vocab/phi_tensor_base_vector.h:69
↓ 42 callersMethodget_dim
(self, axis_name: str)
python/paddle/distributed/fleet/base/topology.py:178
↓ 42 callersFunctiongpc_vertex_create
paddle/phi/kernels/funcs/gpc.cc:690
↓ 42 callersMethodid
(self)
python/paddle/distributed/auto_parallel/static/graph.py:27
↓ 42 callersMethodid
paddle/cinn/operator_fusion/pattern.h:54
↓ 42 callersMethodis_bool
paddle/cinn/common/type.cc:300
↓ 42 callersMethodload
paddle/fluid/pybind/complex.h:51
↓ 42 callersMethodop
(self)
python/paddle/distributed/auto_parallel/static/cost/base_cost.py:687
↓ 42 callersMethodops
paddle/cinn/operator_fusion/pattern.h:56
↓ 42 callersMethodpop_back
paddle/fluid/imperative/op_base.h:254
↓ 42 callersFunctionsize
paddle/utils/span.h:83
↓ 42 callersMethodstart
Start the timer.
python/paddle/distributed/fleet/utils/timer_helper.py:59
↓ 42 callersMethodtensor
paddle/phi/core/meta_tensor.cc:273
↓ 42 callersMethodtensor
paddle/cinn/lang/placeholder.h:63
↓ 42 callersFunctionto_list
(value)
python/paddle/hapi/model.py:82
↓ 41 callersMethodAcquireMemoryFromPrimitive
paddle/phi/backends/onednn/onednn_reuse.h:370
↓ 41 callersFunctionAssign
paddle/fluid/framework/ir/xpu/quant_utils.cc:30
↓ 41 callersMethodIPUPlace
paddle/phi/common/place.h:184
↓ 41 callersFunctionMakeMetaTensor
paddle/phi/api/lib/api_gen_utils.cc:116
↓ 41 callersMethodOpBase
paddle/fluid/framework/new_executor/new_executor_defs.cc:222
↓ 41 callersMethodSetMember
paddle/phi/core/sparse_coo_tensor.cc:134
↓ 41 callersMethodShareDim
paddle/fluid/framework/operator.cc:341
↓ 41 callersFunctionTransToPhiBackend
paddle/phi/core/compat/convert_utils.cc:31
↓ 41 callersMethod_PD_ToPaddleIntArray
paddle/phi/api/include/compat/c10/util/ArrayRef.h:154
↓ 41 callersFunction_as_list
(x)
python/paddle/base/backward.py:2361
↓ 41 callersMethodadd_tile_offset
Advances an iterator along logical dimensions of matrix in units of whole tiles
paddle/phi/kernels/fusion/cutlass/memory_efficient_attention/iterators/predicated_tile_access_iterator_residual_last.h:1909
↓ 41 callersMethodapi
(self, stmt, inputs)
python/paddle/jit/sot/symbolic/interpreter.py:146
↓ 41 callersFunctionconvert_np_dtype_to_dtype_
Convert the data type in numpy to the data type in Paddle. Args: np_dtype (np.dtype|str): The data type in numpy or valid data type
python/paddle/base/framework.py:1487
↓ 41 callersFunctiondynamic_guard
()
test/legacy_test/test_activation_op.py:40
↓ 41 callersMethodend
paddle/fluid/ir_adaptor/translator/program_translator.h:71
↓ 41 callersFunctionfma
paddle/phi/kernels/fusion/gpu/mmha_util.cu.h:2366
↓ 41 callersMethodfrom_numpy
Generates a InputSpec based on the description of input np.ndarray. Args: tensor(Tensor): the source numpy ndarray to ge
python/paddle/static/input.py:320
↓ 41 callersMethodmemory_size
paddle/phi/core/dense_tensor_impl.cc:33
↓ 41 callersMethodpop_back
paddle/pir/src/core/block.cc:40
↓ 41 callersMethodx
paddle/fluid/pir/dialect/operator/ir/manual_onednn_op.h:77
↓ 40 callersMethodAddOutputs
paddle/pir/include/core/operation_utils.h:132
↓ 40 callersFunctionCastPyArg2Value
paddle/fluid/pybind/eager_utils.cc:2327
↓ 40 callersMethodCusparseCall
paddle/phi/backends/gpu/gpu_context.cc:1031
↓ 40 callersFunctionInput
paddle/fluid/framework/operator.h:524
↓ 40 callersMethodRecord
paddle/fluid/pybind/op_callstack_utils.cc:78
↓ 40 callersMethodSetOutput
paddle/fluid/imperative/op_base.h:103
↓ 40 callersFunctionTensorFromVector
paddle/phi/core/tensor_utils.cc:496
↓ 40 callersMethod_get_data
( self, name, shape, dtype, set_feed_dict=True, append_batch_size=True )
test/legacy_test/test_layers.py:651
↓ 40 callersMethodadd
(self, item)
python/paddle/distributed/fleet/meta_parallel/sharding/group_sharded_stage3.py:52
↓ 40 callersMethodapply
(self, main_programs, startup_programs, context=None)
python/paddle/distributed/passes/pass_base.py:91
↓ 40 callersMethodcheck_output_with_place
Check whether the fetched outputs of the origin program and the optimized program are the same. For inference model, the par
test/ir/pass_test.py:121
↓ 40 callersMethodcontains
paddle/fluid/pir/dialect/distributed/ir/dist_attribute.h:52
↓ 40 callersFunctioncreate_test_zero_size_class
(parent)
test/legacy_test/test_activation_op_zero_size.py:63
↓ 40 callersMethodfloat_attr
paddle/pir/src/core/builder.cc:90
↓ 40 callersMethodget_concrete_program
( self, *args: _InputT.args, **kwargs: _InputT.kwargs )
python/paddle/jit/dy2static/program_translator.py:588
↓ 40 callersMethodget_stage_id
(self)
python/paddle/distributed/fleet/base/topology.py:666
↓ 40 callersMethodhas_var
(self, name)
python/paddle/base/framework.py:4603
↓ 40 callersFunctionhash
paddle/pir/include/core/type.h:129
↓ 40 callersMethodis_var
If the node is a variable, then return true. Returns: bool: indicate whether the node is a variable.
python/paddle/base/framework.py:5215
↓ 40 callersMethodlength
paddle/phi/common/pstring.h:281
↓ 40 callersMethodminimize
This function is similar as `optimizer.minimize()`, which performs parameters updating. If the scaled gradients of parameters contai
python/paddle/amp/grad_scaler.py:816
↓ 40 callersMethodname
paddle/fluid/inference/api/demo_ci/custom_relu_pass.cc:21
↓ 40 callersFunctionp_str_to_dict
Parses a strategy string like 'd2·t2' into a config dictionary.
test/flex_checkpoint/test_strategy_conversion.py:24
↓ 40 callersFunctionscale
paddle/fluid/eager/api/generated/eager_generated/forwards/scale.cc:35
↓ 40 callersMethodstep
This function is similar as `optimizer.step()`, which performs parameters updating. If the scaled gradients of parameters contains N
python/paddle/amp/grad_scaler.py:857
↓ 40 callersFunctionunion_free_vars
(*free_vars: dict[str, Any])
python/paddle/jit/sot/opcode_translator/executor/guard.py:150
↓ 39 callersMethodAcquireForwardPrimitiveDescriptor
paddle/phi/backends/onednn/onednn_reuse.h:303
↓ 39 callersMethodClear
paddle/fluid/framework/scope_pool.cc:47
↓ 39 callersMethodElementOf
paddle/cinn/common/type.cc:132
↓ 39 callersMethodGet
* \brief Get an Op for a given operator name. * Will raise an error if the op has not been registered. * @param op_name Name of the operator.
paddle/cinn/hlir/framework/op.h:138
↓ 39 callersMethodGetAllocator
paddle/phi/core/memory/allocation/thread_local_allocator.h:41
↓ 39 callersFunctionGetGpuLaunchConfig1D
According to NVIDIA, if number of threads per block is 64/128/256/512, * cuda performs better. And number of blocks should be greater (at least * 2x
paddle/phi/backends/gpu/gpu_launch_config.h:107
↓ 39 callersMethodGetInputNames
paddle/fluid/inference/api/analysis_predictor.cc:2477
↓ 39 callersMethodGetPlace
paddle/phi/core/distributed/comm_task.h:76
↓ 39 callersMethodInterpretCall
paddle/ap/include/axpr/cps_interpreter.h:66
↓ 39 callersMethodIsMainGraph
paddle/fluid/framework/ir/graph.h:385
↓ 39 callersMethodParse
(self, tree)
python/paddle/incubate/cc/ap/apy_to_axpr_json.py:90
↓ 39 callersMethodSetLoDLevel
test/cpp/fluid/framework/var_type_inference_test.cc:144
↓ 39 callersFunctionast_to_source_code
Transforms ast node into source code.
python/paddle/jit/dy2static/ast_utils.py:21
↓ 39 callersFunctionbuild_input
(input_size, dtypes)
python/paddle/hapi/model_summary.py:514
↓ 39 callersMethodforward_log_det_jacobian
The log of the absolute value of the determinant of the matrix of all first-order partial derivatives of the inverse function. Args:
python/paddle/distribution/transform.py:227
↓ 39 callersFunctionfunc
( var_52, var_54, var_38, var_17, var_57, var_53, var_58, var_56, var_55 )
test/cinn/test_same_input_fusion.py:51
↓ 39 callersFunctionget
Returns a pointer
paddle/phi/kernels/fusion/cutlass/memory_efficient_attention/iterators/predicated_tile_access_iterator_residual_last.h:326
↓ 39 callersMethodget_attr
(self, attrs, idx)
test/ir/inference/dist_llama_inference_model.py:407
↓ 39 callersMethodget_tensor
(name)
test/ir/pir/test_save_load_params.py:44
↓ 39 callersMethodhas_var
(self, var_name)
python/paddle/distributed/fleet/meta_optimizers/sharding/shard.py:58
↓ 39 callersFunctioninit
()
test/cinn/test_same_input_fusion.py:28
↓ 39 callersMethodis_complex
(self)
python/paddle/jit/sot/opcode_translator/executor/variables/basic.py:836
↓ 39 callersMethodmodels_infer
(self, arch, pretrained=False, batch_norm=False)
test/legacy_test/test_vision_models.py:24
↓ 39 callersMethodoffset
paddle/fluid/pir/dialect/kernel/ir/kernel_type.cc:43
↓ 39 callersFunctionpow
(x, y)
test/autograd/utils.py:256
← previousnext →1,301–1,400 of 118,755, ranked by callers