You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

gongweibao f9c97dd728 Add distribution supported (#30578) ... Add distribution supported		4 years ago
..
.gitignore	…
CMakeLists.txt	Ascend rc (#30483)	4 years ago
ascend_wrapper_py.cc	Add distribution supported (#30578)	4 years ago
ascend_wrapper_py.h	Add distribution supported (#30578)	4 years ago
box_helper_py.cc	…
box_helper_py.h	…
communicator_py.cc	Feature/large scale kv save base/delta (#27470)	5 years ago
communicator_py.h	Integrated Trainer of Parameter Server (API add `fluid.contrib.layers.sparse_embedding` only) (#22957)	5 years ago
compatible.cc	check op_version_registry in CI test, test=develop (#28402)	5 years ago
compatible.h	[Pass Compatible] Bind python compatible. (#27262)	5 years ago
const_value.cc	…
const_value.h	…
crypto.cc	Add crypto python (#24836)	5 years ago
crypto.h	Add crypto python (#24836)	5 years ago
data_set_py.cc	Refine error message in pybind folder (#24886)	5 years ago
data_set_py.h	…
exception.cc	Enrich the python error types of paddle & polish format (#28124)	5 years ago
exception.h	…
fleet_py.cc	【Paddle.Fleet】Support local save sparse param (#30175)	4 years ago
fleet_py.h	[Feature] one ps (3/4) (#29604)	5 years ago
fleet_wrapper_py.cc	add topo-aware in heter-ps (#30087)	5 years ago
fleet_wrapper_py.h	…
generator_py.cc	add cuda generator (#26786)	5 years ago
generator_py.h	add cpu random Generator (#26013)	5 years ago
global_value_getter_setter.cc	Add support for mkldnn ops types selection with FLAGS in dygraph (#27482)	5 years ago
global_value_getter_setter.h	…
gloo_context_py.cc	Initialize gloo for low level collective apis (#27672)	5 years ago
gloo_context_py.h	[api 2.0] add collective op for cpu using gloo and paddle.distributed.* apis (#26552)	5 years ago
gloo_wrapper_py.cc	…
gloo_wrapper_py.h	…
heter_wrapper_py.cc	add heter ps mode (#25682)	5 years ago
heter_wrapper_py.h	add heter ps mode (#25682)	5 years ago
imperative.cc	Set expected place in child thread for dataloader to avoid costing cuda memory on other card (#30338)	4 years ago
imperative.h	…
inference_api.cc	[Inference] zero_copy_tensor supports int8_t (#30053)	5 years ago
inference_api.h	…
ir.cc	…
ir.h	…
nccl_wrapper_py.cc	…
nccl_wrapper_py.h	…
op_function.h	add View(reuse allocation) strategy on squeeze, unsqueeze, reshape, flatten op (#29913)	4 years ago
op_function_generator.cc	Fix compilcation on CANN20.1 and older (#30494)	4 years ago
protobuf.cc	add complex64 and complex128 type; add +-*/@ and slice opreator for c… (#29199)	5 years ago
protobuf.h	…
ps_gpu_wrapper_py.cc	Heter ps new (#30198)	4 years ago
ps_gpu_wrapper_py.h	heter box (#29734)	5 years ago
pybind.cc	Add distribution supported (#30578)	4 years ago
pybind_boost_headers.h	…
reader_py.cc	Add pin memory control for BufferedReader (#26026)	5 years ago
reader_py.h	…
tensor_py.h	Set expected place in child thread for dataloader to avoid costing cuda memory on other card (#30338)	4 years ago