Paddle

History

Huihuang Zheng 649868ffb2 [Dy2stat] Fix the bug that loop_body_func may return single element (#31806 ) Our old `loop_body` function may return single element when `loop_vars` just contains only 1 element, which can cause bug. The key point of this PR is forcing `loop_body` functions always return tuple.		4 years ago
..
amp	Add matmtl_v2 to amp list (#28693 )	4 years ago
dygraph_to_static	[Dy2stat] Fix the bug that loop_body_func may return single element (#31806 )	4 years ago
__init__.py	Update the demo code and the doc of varbase.backward. (#26506 )	5 years ago
base.py	Set expected place in child thread for dataloader to avoid costing cuda memory on other card (#30338 )	4 years ago
checkpoint.py	Polish save load en doc details (#27845 )	4 years ago
container.py	fix a bug of Sequential::__getitem__ (#30899 )	4 years ago
io.py	Fix bug for 'save mutiple method' (#30218 )	4 years ago
jit.py	[save/load] Fix bug with input_spec=dict[InputSpec] in jit.save (#31517 )	4 years ago
layer_object_helper.py	add use_mkldnn attribute to ops in dygraph (#25773 )	5 years ago
layers.py	let LayerList could add [None], test=develop (#30911 )	4 years ago
learning_rate_scheduler.py	Upgrade string literals to raw string (#28989 )	4 years ago
math_op_patch.py	add warning message when dtypes of operator are not same (#31136 )	4 years ago
nn.py	Call new cudnn batch norm API regardless of data type and data layout (#30157 )	4 years ago
parallel.py	Remove scale loss before reduce in dygraph (#30807 )	4 years ago
parallel_helper.py	add the paddle.distributed.split api (#29970 )	4 years ago
profiler.py	move imperative to dygraph, test=develop	6 years ago
rnn.py	Upgrade string literals to raw string (#28989 )	4 years ago
static_runner.py	Polish jit.save/load design & remove paddle.SaveLoadConfig (#27623 )	4 years ago
tracer.py	Split train_mode and has_grad for tracer (#29064 )	4 years ago
varbase_patch_methods.py	Remove scale loss before reduce in dygraph (#30807 )	4 years ago