Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/PaddlePaddle/Paddle
/ functions
Functions
118,755 in github.com/PaddlePaddle/Paddle
⨍
Functions
118,755
◇
Types & classes
32,748
↳
Endpoints
291
↓ 24 callers
Function
CanonicalAxis
paddle/phi/kernels/funcs/axis_utils.h:22
↓ 24 callers
Method
CastToAxprValue
paddle/ap/include/drr/drr_value_helper.h:34
↓ 24 callers
Function
DeviceMemoryStatCurrentValue
paddle/phi/core/memory/stats.cc:84
↓ 24 callers
Function
GetKeyFromPlace
paddle/fluid/distributed/collective/common.cc:42
↓ 24 callers
Function
GetLoopExtent
paddle/cinn/ir/schedule/ir_schedule_util.cc:97
↓ 24 callers
Function
GetOpsInPattern
paddle/cinn/operator_fusion/pattern.h:236
↓ 24 callers
Method
GetOutputName
paddle/fluid/prim/utils/static/composite_grad_desc_maker.h:286
↓ 24 callers
Function
GetPartialTensor
paddle/phi/core/distributed/utils.cc:20
↓ 24 callers
Function
GetPirAttributeClass
paddle/ap/src/paddle/pir/attribute_method_class.cc:92
↓ 24 callers
Method
GradOpNum
paddle/fluid/imperative/layer.cc:219
↓ 24 callers
Function
IsComplexType
paddle/fluid/framework/data_type.h:243
↓ 24 callers
Method
Lookup
paddle/cinn/hlir/op/custom_call.cc:65
↓ 24 callers
Function
ParseCommandLineFlags
paddle/common/flags.h:166
↓ 24 callers
Method
PartialInitWithAllocator
paddle/phi/backends/gpu/gpu_context.cc:1151
↓ 24 callers
Method
Prove
paddle/cinn/common/integer_set.cc:176
↓ 24 callers
Method
Query
paddle/phi/backends/event.cc:90
↓ 24 callers
Function
Register
paddle/fluid/pybind/graph.cc:251
↓ 24 callers
Method
RemoveAttr
paddle/fluid/framework/op_desc.cc:682
↓ 24 callers
Method
RenameOutput
paddle/fluid/framework/op_desc.cc:909
↓ 24 callers
Method
ResizeAndAllocate
@jim19930609: This interface will be further modified until we finalized the design for Allocator - Allocation For now, we have to temporarily a
paddle/phi/core/dense_tensor.cc:236
↓ 24 callers
Method
SetOutputType
paddle/fluid/framework/var_type_inference.h:121
↓ 24 callers
Function
SetValue
paddle/phi/core/distributed/auto_parallel/reshard/nd_mesh_reshard_function.h:38
↓ 24 callers
Method
SimpleComputeAt
paddle/cinn/ir/schedule/ir_schedule.cc:462
↓ 24 callers
Function
SizeOfType
paddle/fluid/framework/data_type.cc:95
↓ 24 callers
Function
TopologySortOperations
paddle/fluid/framework/ir/graph_helper.cc:152
↓ 24 callers
Method
UpstreamNodes
paddle/ap/include/graph/node_arena.h:164
↓ 24 callers
Method
VisitUpstreamNodes
paddle/ap/include/drr/drr_graph_descriptor.h:43
↓ 24 callers
Function
_binary_creator_
( method_name, op_type, reverse=False, scalar_method=None )
python/paddle/base/layers/math_op_patch.py:638
↓ 24 callers
Function
_binary_creator_
( method_name, python_api, reverse=False, scalar_method=None, )
python/paddle/pir/math_op_patch.py:522
↓ 24 callers
Method
_parse
(self, text)
test/legacy_test/test_print_tensor_in_gpu.py:449
↓ 24 callers
Function
_to_sequence_except_dict
In this function, dict is not viewed as sequence.
python/paddle/static/nn/control_flow.py:2141
↓ 24 callers
Method
all_var_nodes
Return all variable nodes included in the graph as a set.
python/paddle/base/framework.py:5630
↓ 24 callers
Method
block_dim
paddle/cinn/ir/lowered_func.cc:698
↓ 24 callers
Method
clear
paddle/pir/src/core/region.cc:116
↓ 24 callers
Method
compile
(self, dot_path)
python/paddle/distributed/fleet/base/graphviz.py:87
↓ 24 callers
Function
convert_np_dtype_to_proto_type
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:1415
↓ 24 callers
Method
count
(self)
python/paddle/distributed/launch/context/device.py:45
↓ 24 callers
Method
create
(cls, value)
test/legacy_test/test_restricted_unpickler.py:466
↓ 24 callers
Method
create_local_and_dist_tensor_pair
(self, np_array)
test/auto_parallel/test_api_dist_branch.py:30
↓ 24 callers
Method
create_op_node
Create a operator node in the graph. Args: op_type(str): the type of the operator node. attrs(dict): the att
python/paddle/base/framework.py:5733
↓ 24 callers
Method
create_persistable_node
Create a persistable variable node in the graph. In IrGraph, it can not distinguish between persistable variables and parameters.
python/paddle/base/framework.py:5672
↓ 24 callers
Function
create_tensor
(scope, name, np_data)
test/legacy_test/test_beam_search_op.py:23
↓ 24 callers
Function
create_test_bf16_class
(parent, check_grad=True)
test/legacy_test/test_pool2d_op.py:814
↓ 24 callers
Function
create_test_cudnn_fp16_class
(parent, check_grad=True)
test/legacy_test/test_pool2d_op.py:726
↓ 24 callers
Function
create_test_fp16_class
(parent, check_grad=True)
test/legacy_test/test_pool2d_op.py:770
↓ 24 callers
Method
disable_trt_plugin_fp16
paddle/fluid/inference/tensorrt/engine.h:510
↓ 24 callers
Function
flops
count FLOPs for operation. Args: op_type (str): the type of operation. input_shapes (dict): the shapes of inputs. at
python/paddle/utils/flops.py:27
↓ 24 callers
Method
generate
(self)
python/paddle/jit/sot/symbolic/compile_cache.py:72
↓ 24 callers
Function
get_logger
(log_level, name="root")
python/paddle/distributed/utils/log_utils.py:18
↓ 24 callers
Function
hash_combine
paddle/ap/include/adt/adt.h:216
↓ 24 callers
Method
is_cuda
paddle/phi/api/include/compat/ATen/core/TensorBody.h:417
↓ 24 callers
Function
is_dim_replicate
(mapping)
python/paddle/distributed/auto_parallel/static/utils.py:101
↓ 24 callers
Function
is_support_complex
paddle/phi/common/type_promotion.h:114
↓ 24 callers
Function
loss_fn
(x, label)
test/auto_parallel/pir/while_unittest_pir.py:32
↓ 24 callers
Method
num_elements
paddle/cinn/runtime/buffer.cc:57
↓ 24 callers
Method
outer
()
test/legacy_test/test_pylayer_clear_dataptr.py:527
↓ 24 callers
Function
pairwise_distance
r""" It computes the pairwise distance between two vectors. The distance is calculated by p-order norm: .. math:: \Vert x \Vert
python/paddle/nn/functional/distance.py:34
↓ 24 callers
Function
parseSchema
paddle/phi/api/include/compat/torch/csrc/jit/function_schema_parser.cpp:559
↓ 24 callers
Method
query
Checks if all work currently captured by event has completed. Returns: bool: Whether all work currently captured by eve
python/paddle/device/__init__.py:1213
↓ 24 callers
Method
raw_stream
paddle/phi/backends/xpu/xpu_context.h:53
↓ 24 callers
Method
rend
paddle/phi/api/include/compat/c10/util/ArrayRef.h:96
↓ 24 callers
Method
requires_grad_
Change if autograd should record operations on parameters in this layer. Parameters: requires_grad (bool): whether autog
python/paddle/nn/layer/layers.py:3557
↓ 24 callers
Method
send
paddle/phi/core/distributed/store/tcp_store_libuv.cc:290
↓ 24 callers
Method
setUp
(self)
test/legacy_test/test_compare_complex_op.py:27
↓ 24 callers
Method
step
A wrapper for Optimizer's step function to finish the update operation of the optimizer.
python/paddle/distributed/fleet/meta_parallel/sharding/group_sharded_optimizer_stage2.py:615
↓ 24 callers
Method
strides
paddle/phi/api/include/compat/ATen/core/TensorBody.h:199
↓ 24 callers
Function
tanh
paddle/phi/kernels/funcs/eigen/extensions.h:391
↓ 24 callers
Method
terminate
(self, force=False)
python/paddle/distributed/launch/job/container.py:140
↓ 24 callers
Method
to<int32_t>
paddle/phi/capi/include/wrapper_base.h:432
↓ 24 callers
Method
to_dict
(self)
python/paddle/distributed/auto_parallel/strategy.py:53
↓ 24 callers
Method
to_list
(self)
python/paddle/jit/sot/opcode_translator/executor/variables/iter.py:230
↓ 24 callers
Method
train
(self, use_cinn)
test/ir/pir/cinn/test_cinn_amp.py:34
↓ 24 callers
Method
true_block
(self)
python/paddle/static/nn/control_flow.py:219
↓ 24 callers
Function
unsqueeze
paddle/phi/api/include/compat/ATen/ops/unsqueeze.h:21
↓ 24 callers
Method
wait
(self)
python/paddle/distributed/auto_parallel/ring_attention.py:97
↓ 24 callers
Method
with_dynamic_shape
paddle/fluid/inference/tensorrt/engine.h:511
↓ 23 callers
Function
BlockReduce
paddle/phi/kernels/funcs/top_k_function_cuda.h:337
↓ 23 callers
Function
CreateCast
paddle/fluid/platform/device/ipu/popart_canonicalization/op_builder.cc:134
↓ 23 callers
Function
Empty
paddle/phi/kernels/strings/strings_empty_kernel.h:37
↓ 23 callers
Function
GetGPUDeviceCount
paddle/phi/backends/gpu/cuda/cuda_info.cc:77
↓ 23 callers
Method
GetNodeFromCache
paddle/fluid/framework/ir/fused_attention_pass.h:277
↓ 23 callers
Method
GetOutputPtr
paddle/fluid/prim/utils/static/composite_grad_desc_maker.h:269
↓ 23 callers
Method
GetPDNode
paddle/fluid/framework/ir/graph_pattern_detector.h:1845
↓ 23 callers
Method
GetSingleInputGrad
TODO(Ruting): modify name to GetNullableSingleInputGrad after Large-scale development
paddle/fluid/prim/utils/static/composite_grad_desc_maker.h:108
↓ 23 callers
Function
GetUseOpsForOutput
paddle/fluid/pir/utils/general_functions.cc:229
↓ 23 callers
Function
GetValueIds
paddle/fluid/framework/new_executor/instruction/instruction_util.cc:79
↓ 23 callers
Method
GetXcclComm
paddle/phi/core/distributed/xccl_comm_context.h:37
↓ 23 callers
Method
Has
paddle/ap/include/axpr/attr_map.h:41
↓ 23 callers
Function
HashValue
paddle/phi/common/place.h:110
↓ 23 callers
Method
InsertIntoCache
paddle/fluid/framework/ir/fused_attention_pass.h:285
↓ 23 callers
Function
IsInvalid
NOTE(zhangbo): Some operators of Paddle support optional inputs or outputs, representing whether the input or output exists. In the Pir, whether the v
paddle/fluid/framework/new_executor/pir_adaptor/pir_adaptor_util.h:126
↓ 23 callers
Function
IsOne
paddle/cinn/common/ir_util.h:170
↓ 23 callers
Method
IsPassCompatible
paddle/fluid/framework/op_version_registry.h:363
↓ 23 callers
Function
MakeDistMetaTensor
paddle/phi/api/lib/api_gen_utils.cc:620
↓ 23 callers
Function
MapDims
paddle/phi/infermeta/spmd_rules/flash_attention.cc:41
↓ 23 callers
Method
Match
paddle/cinn/backends/function_prototype.cc:29
↓ 23 callers
Function
NoGrad
paddle/fluid/framework/custom_operator_utils.h:63
↓ 23 callers
Method
RecoverOutputName
paddle/fluid/prim/utils/static/composite_grad_desc_maker.h:499
↓ 23 callers
Method
Run
test/cpp/pir/pass/pass_manager_test.cc:168
↓ 23 callers
Method
Run
test/cpp/pir/custom_engine/fake_cpu_engine.h:147
← previous
next →
1,901–2,000 of 118,755, ranked by callers