提交 · e340df013edaea518ffb509b3e7d801e4a7dac82 · 机器未来 / Paddle

30 8月, 2019 1 次提交
- C
  Support feed single persistable variable to PE (#19417) · e340df01
  由 chengduo 提交于 5年前
```
* update executor feed
```
  未验证
  
  e340df01
29 8月, 2019 3 次提交

L
clean up intel labeled TODOs (#19476) · ba368bf6
由 lidanqing 提交于 5年前
```
test=develop
```
ba368bf6

support debug each output of each ins (#19004) · 1fe468d3

由 Thunderbrook 提交于 5年前

* dump slot

* test

* proto

* dump slot

* test

* proto

* code style

* code style

* code style

* style

* add delete after unseen days

* add unseen days

* code style

* conflict solve
test=develop

* add clear model

* code style
test=develop

* code style
test=develop

* support debug tensor of each ins
test=develop

* support debug tensor of each ins
test=develop

* learning rate

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style
test=develop

* code style
test=develop

* unitest

* style

* style

* multi phase

* add channel

* code style

* style

* style

* unitest

* style

* define

* define
test=develop

* style
test=develop

* rm define
test=develop

* linux

* linux
test=develop

* style
test=develop

* output format
test=develop

* windows ci
test=develop

未验证

1fe468d3

Z
support fc sort by number, test=develop (#19466) · bd35a7f0
由 zhang wenhui 提交于 5年前
```
fleet_desc sort fc name by dictionary sort, but we want to sort by number.
```
未验证

bd35a7f0

28 8月, 2019 5 次提交

fix row_conv_op to force it support lodtensor and tensor input simultaneously,... · 1d0f0431

由 Double_V 提交于 5年前

fix row_conv_op to force it support lodtensor and tensor input simultaneously, test=develop (#19412)

Support Tensor input for row_conv_op

1d0f0431

fix con2d transpose bias by create and init it in build_once (#18968) · 1ce0a09e

由 Jiabin Yang 提交于 5年前

* fix con2d transpose bias by create and init it in build_onee

* fix API spec

* test=develop, invoke ci

* fix bias_attr and act has no effect error on layer norm, conv2dTranpose, billinearTensorProduct, sequece_conv. fix original_mode not used error on GRUunit. fix sample_weight not set error on NCE. Add ut for all thoese layer

* test=develop, change success standard for conv2dTranspose

* test=develop, fix test_layers to invoke some error branch

* test=develop, fix sample code

* test=develop, fix BilinearTensorProduct failed in dygraph mode

* test=develop, fix test_layers segment fault error

未验证

1ce0a09e

Y
adapte fleet api for localsgd and support nccl comm configuration in executor (#19443) · 4ef6b845
由 Yi Liu 提交于 5年前
```
test=develop
```
未验证

4ef6b845

Fix the correctness of async mode at distributed training (#18863) · 65c73684

由 tangwei12 提交于 5年前

* fix correctness of the communicator

* fix a bug in send thread when sending var context is empty, test=develop

* add lookup_table_prefetch_op and prefetch optimize, test=develop

* remove remote prefetch GPU supported

* word2vec force with CPU, test=develop

* test dist remote lookup table force with CPU, test=develop

未验证

65c73684

C
Open test_parallel_dygraph_se_resnext (#19342) · e26411ce
由 chengduo 提交于 5年前
```
* enabel test_parallel_dygraph_se_resnext
test=develop
```
未验证

e26411ce

27 8月, 2019 4 次提交

supports multiple NCCL communicators preserved in NCCLCommContext (#19407) · efb05ba2

由 Yi Liu 提交于 5年前

* supports multiple NCCL communicators preserved in NCCLCommContext
test=develop

* add ut for c_comm_init_all operator and fix cuda resource release problem
test=develop

未验证

efb05ba2

improve sequence_conv api doc (#19316) · a9cd5136

由 Aurelius84 提交于 5年前

* improve sequence_conv api doc test=develop

* add warning for padding param test=develop

modify into deprecated

未验证

a9cd5136

Z
fix fleet_desc bug && support format for abacus hotstart (#19430) · 0d794983
由 zhang wenhui 提交于 5年前
```
fix fleet_desc dense_table unsort bug ，not  support format for abacus hotstart yet.
```
未验证

0d794983

Support Tensor input with padding for warpctc op (#19322) · 482ce818

由 vincentXiyu 提交于 5年前

* support tensor input with padding for warpctc op

* merge with develop

* test=develop

* modified python API examples test=develop

* nn.py is modified for code coverage test=develop

* update documents info about warpctc op in API.spec test=develop

* add test_warpctc_with_padding in test_layers test=develop

* add warning log for cuda_version back to warpctc_op.cc

* modify API.spec for warpctc op test=develop

* modify API.spec

* update warpctc test to new CompiledProgram API test=develop

* modify code examples for warpctc op test=develop

* modify API.spec for warpctc op test=develop

* modify API.spec for warpctc op test=develop

482ce818

26 8月, 2019 11 次提交

C
Fix optimizer bug (#19410) · bfb6ac81
由 chengduo 提交于 5年前
```
* fix optimizer bug
test=develop
```
未验证

bfb6ac81

Fix bug of getting bool Flags from os.environ (#19349) · 6fb310ae

由 Leo Chen 提交于 5年前

* fix bug of getting bool Flags from os.environ, test=develop

* add empty loss_name in CompiledProgram for inplace grad test, test=develop

6fb310ae

Python infer api update and add unit test (#19353) · 32598ffd

由 liu zhengxi 提交于 5年前

* python inference api supports numpy and add unit test, fix unit test fail in test_slim_int8_googlenet and test_slim_int8_mobilenet

未验证

32598ffd

Z

remove recordio convert in dataset, test=develop (#19387) · 807c7a47
由 Zeng Jinle 提交于 5年前

未验证

807c7a47
C
enabel seresnext reduce test (#19341) · 11070cbf
由 chengduo 提交于 5年前
```
test=develop
```
未验证

11070cbf
G
fix- raise io error when user load from non-existed dir test=develop (#19384) · 10643b4e
由 Ghost Under Moon 提交于 5年前
```
This PR fix problem with issue #18096 , which raise an error for user to specify the error about load dir is wrong
```
10643b4e
M
delete recordio writer (#19406) · c2e5eaa2
由 mapingshuo 提交于 5年前
```
test=develop
```
未验证

c2e5eaa2

Lookahead optimizer (#19386) · d5ac87ec

由 mapingshuo 提交于 5年前

* Add lookahead optimizer

* add unittest for lookahead optimizer

test=develop

* add doc string for LookaheadOptimizer

test=develop
test=document_preview

* add API spec for lookahead

test=develop
test=document_preview

* modify api spec

test=develop
test=document_preview

* modified doc string

* modify the test file

test=develop
test=document_preview

* modify doc string

test=develop
test=document_preview

未验证

d5ac87ec

Optimized error reporting information (#19173) · da127d11

由 silingtong123 提交于 5年前

* test=develop,Optimized error reporting information

* test=develop,add importscipy unittest

* test=develop, rename the file and function

da127d11

J
fix problem that get_attr method can't using default mode when we call has_attr in dygraph (#19328) · 55931db4
由 Jiabin Yang 提交于 5年前
```
* add default getItem

* test=develop, fix has_attr disabled error in Layer

* test=develop, fix GroupNorm and deepcf bug on attrs
```
未验证

55931db4
T
fix distribute transpiler GRPC error code 4, RPC Deadline (#18984) · 19dac67e
由 tangwei12 提交于 5年前
```
* fix sync mode hang in transpiler
* remove sync mode in send/recv
* replace PADDLE_ENFORCE with PADDLE_ENFORCE_NE
```
未验证

19dac67e

24 8月, 2019 1 次提交
- Y
  Fix arg do_model_average in param_attr (#19376) · 5d1575cf
  由 Yibing Liu 提交于 5年前
```
* Fix arg do_model_average in param_attr

test=develop

* Update api spec

test=develop
```
  未验证
  
  5d1575cf
23 8月, 2019 3 次提交

Z
add fleet_desc config feature & multi_sparse table, test=develop (#18827) · 4a3c4b8f
由 zhang wenhui 提交于 5年前
```
 add fleet_desc config feature & multi_sparse table,
```
未验证

4a3c4b8f

Update Light-NAS to support latency-aware search (#19050) · 1799c257

由 Jiancheng Li 提交于 5年前

* update light_nas_strategy: add latency constraint

test=develop

* update light_nas_strategy: update get_model_latency

test=develop

* update light_nas_strategy: add more check

test=develop

* update light_nas test

test=develop

* update light_nas test

    test=develop

* minor update light_nas test

    test=develop

* minor update light_nas test

test=develop

* update light_nas test

test=develop

* update _constrain_func of light_nas_strategy

test=develop

* update _constrain_func of light_nas_strategy

test=develop

* remove unused code

test=develop

1799c257

Z
Add the max-pool2d quantization support and the partial quantization support. (#19310) · 0fe72469
由 Zhen Wang 提交于 5年前
```
* add pool2d quantization support, only for max-pooling.

* add the partial quantization support.
```
未验证

0fe72469

22 8月, 2019 8 次提交

L

update inplace grad test to new CompiledProgram API, test=develop (#19359) · d49c2bad
由 Leo Chen 提交于 5年前

d49c2bad

Fix sequence mask in dygraph (#19271) · b2c4f76c

由 Yibing Liu 提交于 5年前

* Fix data parallel & sequence mask in dygraph

test=develop

* Revert change in data_parallel

test=develop

未验证

b2c4f76c

C
Update CompiledProgram (#18919) · 4278518f
由 chengduo 提交于 5年前
```
* use PE for compiler
test=develop
```
未验证

4278518f

翟

Use sparse matrix to implement FusedEmbeddingSeqPoolGradKernel (#19153) · 2e3ee579

由翟飞跃提交于 5年前

* Implement the operator with sprase matrix multiply

* Update the URL of mklml library.

test=develop

* Disable MKLML implematation when using no-linux.

test=develop

* optimize bp with mkl sparse matrix
test=develop

2e3ee579

Enhance OpTest to check the consistency of operators when using and not using inplace (#19101) · a9d5fc51

由 Leo Chen 提交于 5年前

* add pybind interface to get all inplace ops, test=develop

* enhance OpTest to check whether the consistency of operator when using and not using inplace, test=develop

* handle corner cases in op_test, test=develop

* support outputs without tensor holder_, like XShape in reshape_op, test=develop

* fix bug, some op has GradOpMaker, but actually no grad_op in OpInfoMap, test=develop

* use reshape_grad instead of reshape in FlattenGradOp, test=develop

* fix error debug dims info for variables like XShape, test=develop

* change computational order in sum_op to relieve computation difference using inplace, test=develop

* add inplace_atol to check group_norm, and skip inplace_grad for mkldnn, test=develop

* follow sneaxiy's comments, test=develop

* remove unused DefaultGradOpDescMaker in mkldnn op, test=develop

a9d5fc51

Supports diagonal initialization in uniform_random op (#19299) · 0d29cf18

由 Aurelius84 提交于 5年前

* add diag init in Uniform_random op test=develop

* modify api.spec test=develop

* fix unform_batch_size_like maker test=develop

* add diag_num and diag_step assert check test=develop

未验证

0d29cf18

C
[Speedup] Make dygraph data parallel faster (#19280) · 5a579df9
由 chengduo 提交于 5年前
```
* update parallel.py
test=develop
```
未验证

5a579df9

Split test_parallel_executor_seresnext to three unit test (#19239) · 6a163231

由 chengduo 提交于 5年前

* increase test_parallel_executor_seresnext time limit
test=develop

* split test_parallel_executor_seresnext
test=develop

* temporally disable reduce_and_allreduce test because of the random failure.
test=develop

* split gpu and cpu
test=develop

未验证

6a163231

21 8月, 2019 2 次提交
- Z
  
  remove is_mem_optimized in Program, test=develop (#19307) · 561232c2
  由 Zeng Jinle 提交于 5年前
  
  未验证
  
  561232c2
- L
  Add elementwise_mul_mkldnn UT with [conv + elt_mul + conv] (#19191) · 3fdecc19
  由 lidanqing 提交于 5年前
```
* add elementwise_mul_mkldnn UT with [conv + elt_mul + conv] to cover avx512=True branch
test=develop

* change a typo.
test=develop
```
  3fdecc19
20 8月, 2019 2 次提交
- X
  fix yolo_box python example (#18925) · 62facc7e
  由 xiaoting 提交于 5年前
```
test=develop, test=document_preview
```
  62facc7e
- D
  distribute launch : add use_paddlecloud argument (#19273) · 0865b5a9
  由 danleifeng 提交于 5年前
```
distribute launch : add use_paddlecloud argument 
```
  0865b5a9

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致