Paddle

History

liuyuhui 87197f8c2e [kunlun]fix sync in multi kunlun xpu dygraph training. (#30943 )		4 years ago
..
contrib	Use correct master weights in AdamW. (#30895 )	4 years ago
dataloader	remove numpy array check in single-process dataloader. test=develop (#30861 )	4 years ago
distributed	Upgrade string literals to raw string (#28989 )	5 years ago
dygraph	fix a bug of Sequential::__getitem__ (#30899 )	4 years ago
incubate	【Paddle.Fleet】Fix tensor table (#30075 )	4 years ago
inference	…
layers	Call new cudnn batch norm API regardless of data type and data layout (#30157 )	4 years ago
tests	[kunlun]fix sync in multi kunlun xpu dygraph training. (#30943 )	4 years ago
transpiler	…
.gitignore	…
__init__.py	[Feature] one ps (3/4) (#29604 )	5 years ago
average.py	…
backward.py	Recompute Offload (#30233 )	4 years ago
clip.py	Upgrade string literals to raw string (#28989 )	5 years ago
communicator.py	[Feature] one ps (3/4) (#29604 )	5 years ago
compiler.py	refine the paddle place support using str (#28769 )	4 years ago
core.py	Separate AVX and NO_AVX compilation, enhance installation error message (#30413 )	4 years ago
data.py	Add static mode check on data() (#27495 )	5 years ago
data_feed_desc.py	…
data_feeder.py	polish printing dtype (#30682 )	4 years ago
dataset.py	add set_trainer_num api in dataset (#29133 )	5 years ago
debugger.py	…
default_scope_funcs.py	…
device_worker.py	enable pipeline to run with Executor.run() (#28373 )	5 years ago
distribute_lookup_table.py	…
dygraph_utils.py	…
entry_attr.py	…
evaluator.py	…
executor.py	refine the paddle place support using str (#28769 )	4 years ago
framework.py	[Kunlun] add gen_bkcl_id_op, support multi XPU cards training using multiprocess (#30858 )	4 years ago
generator.py	refine the paddle place support using str (#28769 )	4 years ago
graphviz.py	…
initializer.py	Upgrade string literals to raw string (#28989 )	5 years ago
input.py	fix gpu outofrange (#29238 )	5 years ago
install_check.py	Reimplement paddle.utils.install_check. (#27771 )	5 years ago
io.py	Supplement PR29988(https://github.com/PaddlePaddle/Paddle/pull/29988 ) (#30507 )	4 years ago
layer_helper.py	…
layer_helper_base.py	Upgrade string literals to raw string (#28989 )	5 years ago
lod_tensor.py	…
log_helper.py	…
metrics.py	Upgrade string literals to raw string (#28989 )	5 years ago
multiprocess_utils.py	…
net_drawer.py	…
nets.py	Upgrade string literals to raw string (#28989 )	5 years ago
op.py	…
optimizer.py	Recompute Offload: fixed bug in memcpy (#30484 )	4 years ago
parallel_executor.py	Add static_only decorator for static apis (#29015 )	5 years ago
param_attr.py	[paddle v2.0.0rc1: API fixs] assign/conv2d/conv2d_transpose/cast/ParamAttr (#29171 )	5 years ago
profiler.py	…
reader.py	Set expected place in child thread for dataloader to avoid costing cuda memory on other card (#30338 )	4 years ago
regularizer.py	Upgrade string literals to raw string (#28989 )	5 years ago
trainer_desc.py	heter box (#29734 )	5 years ago
trainer_factory.py	heter box (#29734 )	5 years ago
unique_name.py	modify doc for unique_name.guard unique_name.generate unique_name.switch, test=develop (#27605 )	5 years ago
wrapped_decorator.py	…