21138eb12a
[DOC][PYTHON] Fix api comment of paddle.fluid.clip.GradientClipByValue ( #17530 )
...
* BugFix: fix api comment of paddle.fluid.clip.GradientClipByValue
* test=develop, test=document_preview
6 years ago
6724a652f3
add __str__ method for tensor and lodtensor to support print test=dev… ( #17588 )
...
* add __str__ method for tensor and lodtensor to support print test=develop
6 years ago
887a39f050
Fix dygraph unique name bug ( #17592 )
...
* fix unique_name growth bug in dygraph mode,test=develop
* change generate_tmp to generate_with_ignorable_key,test=develop
6 years ago
e9216d0602
Enable logical operators for the nGraph Bridge. ( #17543 )
...
test=develop
6 years ago
cbaf9e5344
Fix api example [ lstm, sequence_enumerate, sequence_expand,sequence_expand_as ] ( #17210 )
...
* fix example; test=develop
* fix api spec; test=develop
* fix api spec; test=develop
* add doc check
test=develop
test=document_preview
* test=develop,test=document_preview
add blank line to fix format, add one more "import"
* fix bug; test=develop
* fix bug; test=develop
6 years ago
e53119f5b1
Fix decayed adagrad example ( #17390 )
...
* fix DecayedAdagrad example;
test=develop
test=document_preview
* add space; test=develop
6 years ago
326bf8291a
add Run Prepared Ctx ( #17616 )
...
add Run Prepared Ctx, fix pybind problem
6 years ago
2a7b321110
Fix the example code in some Python API. ( #17343 )
...
* Fix the example code in some Python API.
test=develop
* Fix the example code in some Python API by adding import.
test=develop
6 years ago
b5f4d5ed0e
Add broadcast operators ( #17503 )
...
* This PR adds broadcast for multi-process. And it could be used in dynamic graph to broadcast parameters.
6 years ago
2280f185d7
BuildStrategy api comment ( #17348 )
...
Python examples of fluid.layers.io.double_buffer and some BuildStrategy's methods.
6 years ago
3db9c8c982
refine shape and split test. test=develop ( #17545 )
6 years ago
0d4cbdad91
[NGraph] Enable elementwise mul operator ( #17552 )
6 years ago
f2694e122d
[NGraph] Enable assign operator for a ngraph, test=develop ( #17437 )
...
* Enable assign operator for a ngraph, test=develop
* Cross_entropy operators needs to be updated
6 years ago
cf02cb5e98
Enable elementwise sub operator for ngraph ( #17527 )
6 years ago
7f8bc49d00
polish_executor_and_add_ctx_cache ( #17536 )
...
* polish_executor_and_add_ctx_cache
6 years ago
8818c94cfb
test=develop ( #17207 )
6 years ago
ca03f4989a
fix distributed launch.py ( #17571 )
...
test=develop
6 years ago
6e11f97708
Add exponential moving average ( #17562 )
...
* Add exponential moving average
test=develop, test=document_preview
* Polish documents
test=develop, test=document_preview
* Update API spec
test=develop, test=document_preview
6 years ago
3ee3611aa7
test=develop, fix test_imperative_resnet failed on CI ( #17583 )
6 years ago
035771512d
Enable elementwise min operator for ngraph ( #17521 )
6 years ago
d817263c80
add unittest of dygraph RL models. ( #17550 )
...
* test=develop
* test=develop
6 years ago
cf60e5a2db
fix API python example ( #17226 )
...
* fix api example. test=develop
* fix API.spec. test=develop
* fix spectral_norm format. test=develpp
* merge develop
* add import. test=develop
* fix indent. test=develop
* fix indent. test=develop
* add import fluid. test=develop
6 years ago
92e7d5d7cc
fix distribute doc test=develop ( #17318 )
...
* fix distribute doc
6 years ago
58f7695ab2
Async exe support communicator ( #17386 )
...
Async exe support communicator
6 years ago
109b5aed5a
[NGraph] Enable reshape operator test=develop ( #17512 )
6 years ago
9bb6a421e3
fix bpr_loss data_norm teacher_student_sigmoid_loss api & fix continuous_value_model ( #17331 )
...
* fix bpr data_norm teacher_student_sigmoid , test=develop test=document_preview
修复了bpr_loss data_norm teacher_student_sigmoid_loss三个api, 同时修复了continuous_value_model文档英文拼写错误
6 years ago
300bd7504d
fix api-doc related bugs test=develop test=document_preview ( #17360 )
...
* fix api doc according to the reviewer's comment test=develop
6 years ago
3bd14263f5
decrease the train loop number to avoid run too long to fail the ci process test=develop ( #17567 )
6 years ago
43d15b9d96
Enable square operator for the nGraph Bridge. ( #17551 )
...
test=develop
6 years ago
ff5fdc0b67
test=develop, fix con2d with no bias ( #17516 )
...
* test=develop, fix con2d with no bias
* test=develop, fix conv_2d with no bias will have null shape
6 years ago
f86f49e779
[NGraph] add increment op to ngraph engine ( #16929 )
...
* add increment op to ngraph engine
test=develop
* fix style errors
test=develop
6 years ago
cf5d271c5a
Fix examples of fluid.layers.sums and fluid.layers.DynamicRNN ( #17308 )
...
* Fix examples of fluid.layers.sums.
test=document_preview
* Correct the example of DynamicRNN and its functions.
test=develop
* Add 'import paddle.fluid as fluid' to examples.
test=develop
* Update API.spec.
test=develop
* Add space lines.
test=develop
* Update the API.spec.
test=develop
6 years ago
2281ebf0f3
Enable the convolution/relu6(bounded_relu) fusion for FP32 on Intel platform. ( #17130 )
...
* Relu6 is the bottleneck op for Mobilenet-v2. As the mkldnn supports the conv/relu6 fusion, we implement it fusion via cpass way. Due to the int8 enabling for this fusion will be supported in MKLDNN v0.20, so this PR is focused on the fp32 optimization.
Below table shows the benchmark(FPS) which measured on skx-8180(28 cores)
Batch size | with fusion | without fusion
-- | -- | --
1 | 214.7 | 53.4
50 | 1219.727 | 137.280
test=develop
* Fix the format issue
test=develop
* Add the missing nolint comments.
test=develop
* Fix the typos.
test=develop
* Register the conv_brelu_mkldnn_fuse_pass for the MKLDNN engine.
test=develop
* Adjust the indentation.
test=develop
* Add the test_conv_brelu_mkldnn_fuse_pass case.
test=develop
* Slightly update the code per Baidu comments.
Let the parameter definition embedded into the code.
That's will make the code easy to understand.
test=develop
6 years ago
3398f99608
Adding AddQuantDequantPass for TensorRT int8 ( #17529 )
...
* add quant_dequant_pass, test=develop
* Add quant_dequant before some ops, such as the elementwise_add op. This is required by TensorRT. test=develop
6 years ago
f9796b1249
Add LAMB Optimizer support ( #17489 )
...
* Add LAMB optimizer
* Expose LAMB Optimizer's APIs
test=develop, test=document_preview
* Cleanup code & doc
test=develop, test=document_preview
* Update lamb optimizer's formula
test=develop
6 years ago
99ab57123c
Enabled ngraph elementwise max operator ( #17517 )
6 years ago
3d19f44a89
remove unused SERIAL compiler option ( #17500 )
...
test=develop
6 years ago
dfdcd91869
Add api doc code examples ( #17285 )
...
* Add api doc code examples
add or fix topk, squeeze, stack, StaticRNN,
StaticRNN memory in doc
test=develop
Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>
* Add squeeze md5.
test=develop
Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>
* Add import package
test=develop
Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>
6 years ago
d2682a84ca
test=develop, fix dygraph batch_norm ( #17465 )
6 years ago
1eb151752e
Enable abs operator for a ngraph test=develop ( #17436 )
6 years ago
ff7f911b4d
add quant_dequant_moving_avg_max_abs op ( #17480 )
...
* add quant_dequant_moving_avg_max_abs op
test=develop
* add more note for quantdequant op
test=develop
6 years ago
306eadcd39
fix eval mode bug; test=develop ( #17499 )
6 years ago
287de41c04
Optimize communicator flags ( #17494 )
...
* optimize communicator flag
* change flags in init py test=develop
6 years ago
9eb19df266
fix uniform_random op,test=develop ( #17492 )
6 years ago
c3949f5699
remove two useless flags: enable_subgraph_optimize, memory_optimize_debug, test=develop ( #17491 )
6 years ago
f82e4d75e7
improve the doc of paddle.fluid.memory_optimize, test=develop ( #17473 )
...
* improve the doc of paddle.fluid.memory_optimize, test=develop
* fix typo, test=develop
6 years ago
65dd7ec2d6
add clear ops in dygraph optimizers,test=develop ( #17484 )
6 years ago
10b23a72c1
Double backward elementwise div ( #17416 )
...
* double backward, elementwise_div
* fix dx empty. test=develop
* bug fix (#17392 )
fix secure bug
* Eanble stack operator for a Ngraph, test=develop (#17406 )
* fix sqrt_grad_grad unittest. test=develop (#17410 )
* fix sqrt_grad_grad unittest. test=develop
* disable sqrt_grad_grad unittest. test=develop
* test=develop, fix unittest
* test=develop, fix unittest
* test=develop, fix unittest
* test=develop, fix bug
* fix unittest. test=develop
* fix unittest dx. test=develop
* tmp fix! for test... test=develop
* reduce tmp, test=develop
* test=develop, reduce tmp
* fix broadcast unittest. test=develop
* fix format. test=develop
* refine code. test=develop
* refine code. test=develop
* refine GetDoubleGradSafeTensor. test=develop
* fix format. test=develop
6 years ago
14f223624f
fix sqrt unittest. test=develop ( #17440 )
6 years ago
977e9fcb27
support elementwise_sub double backward ( #17476 )
...
add elementwise_sub_grad_grad op for backward of backward calculation
6 years ago
75cda4d9df
fix data_feed_desc.py example run error ( #17452 )
...
* fix data_feed_desc.py example run error
test=develop
test=test=document_preview
* fix data_feed_desc.py example display error
test=develop
test=document_preview
* update API.spec for DataFeedDesc
test=develop
test=document_preview
6 years ago
e32f4c4f9d
examples use code-block in dataset.py ( #17451 )
...
* examples use code-block in dataset.py
test=develop
test=document_preview
* add QueueDataset example
test=develop
test=document_preview
6 years ago
05df39ac06
support sparse table get shard_num from TableParameter ( #17443 )
...
test=develop
6 years ago
0217555530
polish parallel dygraph code ( #17164 )
...
* add var grad hook test=develop
6 years ago
15453d05a8
test=develop, fix AdgradOptimizer example code ( #17401 )
6 years ago
6a53fa95e7
improve the API Sample of DataFeeder, memory_optimize and release_memory ( #17374 )
...
* improve the API Sample of DataFeeder, memory_optimize and release_memory, test=develop
* update API.spec, test=develop, test=document_preview
* tweak the code format of feed API, test=develop
* update API.spec, test=develop
* improve doc for DataFeeder and default_main_program, test=develop
6 years ago
712bfb17cb
fix recurrent_op,test=develop ( #17433 )
6 years ago
e336dc86bb
[Speed] Refine the Executor when the num_thread=1 ( #17405 )
...
Refine the Executor when the num_thread=1
6 years ago
30e178fa2c
init auto loss scaling ( #17194 )
...
* init auto loss scaling
test=develop
* change API.spec
* change ifelse to switch and use reduce_sum to optimize checking isfinite
test=develop
* Remove redundant code
test=develop
6 years ago
58d5c61a29
fix sqrt_grad_grad unittest. test=develop ( #17410 )
...
* fix sqrt_grad_grad unittest. test=develop
* disable sqrt_grad_grad unittest. test=develop
6 years ago
6ee6700fac
Eanble stack operator for a Ngraph, test=develop ( #17406 )
6 years ago
34369944f5
support config file, cvm, load, save, shrink ( #17319 )
...
* support config file, cvm, load, save, shrink
test=develop
* fix error of worker_num & add table.compress_in_save
test=develop
* fix code style
test=develop
* fix save model bug
test=develop
6 years ago
66d51206b1
add save/load model, shrink table, cvm, config file & fix pull dense bug ( #17118 )
...
* add save/load model, shrink table, cvm, config file & fix pull dense bug
test=develop
* fix global shuffle bug, fix pull dense bug, fix release memeory bug, fix shrink error
add client flush, add get data size
test=develop
* fix global shuffle bug
test=develop
* fix global shuffle bug
test=develop
* fix code style
test=develop
* fix code style & modify pslib cmake
test=develop
* fix error of _role_maker
test=develop
* fix code style
test=develop
* fix code style
test=develop
* fix code style
test=develop
* fix code style
test=develop
* fix code style
test=develop
* fix windows compile error of fleet
test=develop
* fix global shuffle bug
* add comment
test=develop
* update pslib.cmake
test=develop
* fix fill sparse bug
test=develop
* fix push sparse bug
test=develop
6 years ago
266444b8af
fix dist launch script test=develop ( #17404 )
6 years ago
1ce7b45b9e
NGraph Added fill_zeros_like op test=develop ( #17295 )
6 years ago
910196524d
NGraph Added dropout and dropout_grad to ngraph test=develop ( #17320 )
6 years ago
b189480734
Ngraph Enable gather operator test=develop ( #17296 )
6 years ago
4ef631013c
Double backward sqrt ( #17387 )
...
* double backward sqrt
* refine unittest. test=develop
* refine test. test=develop
* remove alpha in unittest. test=develop
6 years ago
829fcc98fb
Fix some APIs' example
...
* test=develop
* test=develop
* test=develop
6 years ago
eab34b2df6
fix_dygraph_mem_leak, test=develop ( #17396 )
6 years ago
5d1ac41b00
Double backward reduce mean ( #17372 )
...
* test=develop, double backward reduce_mean
* add comment. test=develop
* fix format. test=develop
* rename GradGrad -> DoubleGrad. test=develop
* fix op_use_default_grad_op_maker.spec. test=develop
6 years ago
d915a04907
Add examples for AdamaxOptimizer ( #17381 )
...
test=develop
6 years ago
bd9bef5a4e
add elementwise_add_grad_grad op ( #17366 )
...
* add elementwise_add_grad_grad op. test=develop
* use defined GradMaker. test=develop
6 years ago
1c6d064627
add collect fpn proposals op,test=develop ( #16074 )
...
* add collect fpn proposals op,test=develop
6 years ago
60be66e2c0
support fc_op double grad ( #17317 )
...
* add double grad for mul_op. test=develop
* fix format. test=develop
* fix format. test=develop
* fix format. test=develop
* refine code. test=develop
* remove setzero. test=develop
* fix dx/dy init bug. test=develop
* fix format. test=develop
6 years ago
ad8bbe587e
Fix some api example codes' bugs and these APIs include load_inference_model, load_vars, save_vars, L1DecayRegularizer and L2DecayRegularizer. ( #17324 )
...
* fix some api example codes' bugs.
* update API.spec. test=develop test=document_preview
* add import fluid. test=develop test=document_preview
6 years ago
2ff7ea3337
Expose sign op ( #17117 )
...
* test=develop
add sign op
* Update nn.py
test=develop
delete stop_gradient assignment
6 years ago
4624d7c642
test=develop, add gradient sort backward strategy ( #17125 )
...
* test=develop, add gradient sort backward strategy
* test=develop, fix test by add FLAGS_cudnn_deterministic on new tests
6 years ago
1d0ba5e815
Fix the example code in some Python API ( #17333 )
...
* Fix the example code in some Python API
* Update paddle/fluid/API.spec
* Fix some examples format
6 years ago
c843e64cf5
Revert "rename the default version from '0.0.0' to 'latest' ( #17304 )" ( #17356 )
...
This reverts commit f456c8beb8
6 years ago
8bae8590ac
add double grad for elementwise_mul op ( #17255 )
...
* add double grad for elementwise_mul. test=develop
* remove comment. test=develop
* fix grad sum. test=develop
* fix for axis expand. test=develop
* add test for axis expand. test=develop
6 years ago
11d3a38f25
add double grad for square op ( #17173 )
...
* add double grad for square. test=develop
* formax code. test=develop
* fix for grad sum. test=develop
* refine shape. test=develop
* refine extract. test=develop
6 years ago
31536016ea
test=develop, test=document_preview, fix 13 api doc and code ( #17293 )
...
* test=develop, test=document_preview, fix all 13 api doc and code
* test=develop, fix rst
* test=develop, refresh API.spec
6 years ago
bc833945a4
Add DropLocalExeScopes in ParallelExecutor ( #17297 )
...
* reset drop local scope counter
test=develop
6 years ago
d4b67e1692
Add Where Op( #16793 )
6 years ago
1bfff02047
Add Diag Op( #17027 )
6 years ago
ddb24d48c5
test=develop ( #17322 )
6 years ago
e32c9888f5
Double backward of conv2d. ( #17211 )
...
* Add conv2d_grad_grad_op
* Extracte the cuDNN conv algo searching code in conv_cudnn_helper.h.
- Now use it in conv2d_grad_grad.
- Will simply the searching code in conv2d and conv2d_grad in next PR.
* Enhance and fix bug in unit testing of gradient_checker.
* Support to fetch empty variables，return None in Python.
6 years ago
f456c8beb8
rename the default version from '0.0.0' to 'latest' ( #17304 )
...
* rename the default version from '0.0.0' to 'latest'
6 years ago
50ad9046c9
add import, test=develop ( #17229 )
6 years ago
4292bd8687
Mod floordiv ( #17251 )
...
* test=develop
add elementwise_mod and elementwise_floordiv, fix equation problem in elementwise_mod
6 years ago
565d309501
Reformat fleet API ( #17135 )
...
* fix some logic in distributed transpiler, test=develop
* reformat fleet API, test=develop
6 years ago
a88a1faa48
Format file path ( #17280 )
...
The parameter dirpath will be passed directly to c++ operater. The file address format will be different under win and UNIX.
6 years ago
5d6a1fcf16
fix infer_from_dataset and train_from_dataset ( #17243 )
...
* fix train_from_dataset and infer_from_dataset example
* add inductive dim for data_reader, example: shape=[-1, 1], then -1 will be inducted through run-time reading of number of elements
6 years ago
2c4462711f
Fix API example code of save_inference_model ( #17274 )
...
* Fix API example code of save_inference_model
test=develop
* Add "import" in exmaple of save_inference_model
* Fix typo "exsample" -> "example"
test=develop
6 years ago
9ed4aaada4
modified formula for Lrn ( #17281 )
...
* modified formula for lrn
test=develop
* modified api.spec
test=develop
6 years ago
d8af44a5a3
test=develop, fix error with training and test on diff device ( #17276 )
6 years ago
e388a1fb66
Repair api example ( #17221 )
...
Fix the following API examples:
paddle.fluid.scope_guard
paddle.fluid.backward.append_backward
paddle.fluid.cpu_places
paddle.fluid.cuda_pinned_places
paddle.fluid.cuda_places
paddle.fluid.in_dygraph_mode
paddle.fluid.CUDAPlace
paddle.fluid.CPUPlace
paddle.fluid.CUDAPinnedPlace
6 years ago
6d1d7c8acf
Fix the KL algorithm bug when calculated the size of tensor. ( #17198 )
...
* resolve #17184
Fix the KL algorithm bug when calculated the size of tensor.
* Add the missing test=develop.
test=develop
6 years ago
dab71e8d97
Fix api example ( #17231 )
...
* fix API examples, test=develop
6 years ago
7bd1d03ee5
Adding lrn op for ngraph engine ( #17189 )
...
* added lrn op test=develop
* Added CreateConstant method test=develop
* avoid duplicates test=develop
6 years ago
65541d83b0
add scale pass for calculating the output scales.test=develop ( #17259 )
6 years ago
8f534696b7
Polish Executor and Compiler doc ( #17262 )
...
* polish doc
test=develop
* updata parallel executor doc
test=develop
* update API.spec
test=develop
* polish code
test=develop
6 years ago
91784f8ec3
Fix code in document. ( #17237 )
6 years ago
648320bb6c
Fix some data and reader related API code ( #17202 )
...
* Fix data and reader related api doc
* Fix data and reader related api doc
Review and fix the example code in some reader related API doc.
These APIs are:
Fix existing API example codes:
paddle.fluid.io.PyReader
paddle.fluid.layers.batch
paddle.fluid.layers.data
paddle.fluid.layers.Preprocessor
paddle.fluid.layers.py_reader
paddle.fluid.program_guard
Add new example codes:
paddle.fluid.io.PyReader.decorate_batch_generator
paddle.fluid.io.PyReader.decorate_sample_generator
paddle.fluid.io.PyReader.decorate_sample_list_generator
paddle.fluid.io.PyReader.reset
paddle.fluid.io.PyReader.start
test=develop
* Add changes to API.spec after changing doc.
test=develop
* Add blanks after python example code
test=develop
* Add blank line at py_reader example code
test=develop
* Merge API.spec
test=develop
* Modify reader.py based on reviewer's comment
test=develop
* Modify API.spec after changing doc
test=develop
* Change reader.py based on reviewer's comment
* Modify example code of decorate_sample_generator
test=develop
* Fix example code of PyReader based on reviewer
test=develop
6 years ago
f2fa3f7300
fix api doc,test=develop ( #17241 )
6 years ago
4f8594088d
Enhance inplace/mem-opt pass and enhance softmax_with_cross_entropy op inplace ( #17225 )
...
* add use_cuda to inplace pass,test=develop
* add test softmax_with_xe_inplace test,test=develop
* fix potential inplace bug
test=develop
* add more skip vars in mem opt pass,test=develop
* follow comment,test=develop
* follow comments,move duplicate out arg check to program->graph,test=develop
6 years ago
8b62f53719
fix nn.py lack of `import warnings` ( #17249 )
...
test=develop
6 years ago
e782b54b9c
update sofmax with axis arg test=develop ( #17190 )
6 years ago
71f0c6d5bd
fix api doc of hash, relu, concat, argmin, argmax, argsoft and all activations ( #17235 )
...
* fix api doc of hash, relu, concat, argmin, argmax, argsoft and all activations funcs with no attrs
test=develop
* refine doc example code
test=develop
* remove >>> in doc example
test=develop
* refine python code block
test=develop
* update API spec
test=develop
6 years ago
ff1661f12a
remove unused FLAGS_warpctc_dir ( #17162 )
...
* remove unused FLAGS_warpctc_dir
test=develop
* remove FLAGS_warpctc_dir
test=develop
6 years ago
a71d8fdb87
Softmax_cross_entropy op add axis ( #16806 )
...
* add attr axis infershape. test=develop
* add CUDA kernel. test=develop
* fix unittest. test=develop
* fix unittest for soft_label. test=develop
* fix fp16 unittest. test=develop
* remove comment code. test=develop
* refine test for axis. test=develop
* add python api. test=develop
* fix doc. test=develop
* fix fp16 unittest. test=develop
* fix ngraph test. test=develop
* fix ENFORCE for test_imperative_transformer. test=develop
* fit for ngraph test. test=develop
* fix after rebase develop. test=develop
* fix doc. test=develop
* fix API.spec. test=develop
* fix test_layers. test=develop
* fix format. test=develop
6 years ago
a40121e4c8
fix the initialization process error. test=develop ( #17213 )
6 years ago
a914d9b116
Quant output scale ( #17215 )
...
* Add MovingAverageAbsMaxScale operator which is only used for calculating the quantization scale.
* test=develop
* change the output into inplace. test=develop
* Revert "test=develop"
This reverts commit 696cf62699ba1e1c98f61f7345ac7060010eb29a.
* Revert "change the output into inplace. test=develop"
This reverts commit a19acd20f07eee82622701a3015e6e9c073a5e0b.
* test=develop.
* update the MovingAverageAbsMaxScaleOp test. test=develop
6 years ago
16922e0093
fix api_example of tree_conv ( #17239 )
...
test=develop
6 years ago
ef66baedc0
Refine api doc ( #17230 )
...
* refine api comment, test=develop
6 years ago
6b0f27e802
Fix some APIs' example ( #17214 )
6 years ago
5817077c99
Fix unexecutable API examples ( #17218 )
...
* fix unexecutable API comments, test=develop
* add API.spec,test=develop
6 years ago
cc95a7516c
fix distribute fpn proposals, test=develop ( #16152 )
...
* fix distribute fpn proposals, test=develop
6 years ago
9ec4615deb
fix profiler and name_scope API examples ( #17212 )
...
* fix profiler and name_scope API examples
test=develop
* update API.spec
test=develop
6 years ago
06325c4de1
disable the test_distillation_strategy temporarily ( #17227 )
...
test_distillation_strategy always failed on a machine with 4 gpus only, disable temporarily and need to figure out the root cause and add it back later
6 years ago
ee2028a110
Add use_cuda to inplace pass ( #17205 )
...
* add use_cuda to inplace pass,test=develop
* add test softmax_with_xe_inplace test,test=develop
6 years ago
f2db475a3e
update ParallelExecutor ( #17204 )
...
test=develop
6 years ago
a72907bbf4
Enhance concat op to support empty input. ( #17015 )
...
* enhance_concat, test=develop
6 years ago
83c4f7721f
use two GPUs to run the exclusive test test=develop ( #17187 )
6 years ago
8092c40560
Modify test timeout ( #17181 )
...
* test=develop
* test=deelop
6 years ago
f938ccec62
remove async executor python api to fix document ( #17174 )
...
* remove async executor python api
test=develop
* remove test_async_executor.py
add executor train_from_dataset demo
test=develop
* fix import bug
test=develop
6 years ago
5dfe2ab9e8
Fix mem leak when converting Tensor to numpy array ( #17182 )
...
* fix mem leak when converting Tensor to numpy array
test=develop
* remove unused unittest,test=develop
* follow comments, test=develop
* fix dygraph bug,test=develop
6 years ago
4e1bc6e805
Rewrite inplace pass and fix gc bug ( #17126 )
...
* fix op graph view
test=develop
* rewrite inplace pass and fix reference count pass bug
test=develop
* fix unittest failed
test=develop
* follow comments, test=develop
6 years ago
aa5307ce5b
fix python3 run_time_error in ops. test=develop ( #17170 )
...
fix python3 run_time_error in layers.ops caused by locals()
6 years ago
e4a52e082f
resolve #17159 ( #17172 )
...
Update the folder name generation mechanism for saving the quantized model and weights.
The folder name would be unique by adding the timestamp postfix.
test=develop
6 years ago
bc48453b73
polish the label_smooth ( #17138 )
...
* polish the label_smooth
test=develop
* polish code
test=develop
6 years ago
626922d3cd
fix run_time_error in uniform_random. test=develop ( #17152 )
...
fix runtimeerror : dictionary changed size during iteration when calling uniform_random in python3+
6 years ago
deb510d451
cvm op feature ( #17081 )
...
cvm without LoD.
6 years ago
554d3a71d2
test=develop fix bug: fix selected_indices in nms ( #17140 )
6 years ago
28d69d710a
Refine dropout gpu memory ( #17095 )
...
* refine_dropout_mem,test=develop
* # This is a combination of 14 commits.
# The first commit's message is:
remove ut test_dist_word2vec in mac ci, will fix it in private, test=develop (#17066 )
# This is the 2nd commit message:
Fleet unify distributed training (#16791 )
* implement distributed transpiler with fleet
# This is the 3rd commit message:
ParallelDyGraph with GPU collective mode (#16827 )
implement dygraph.parallel.DataParallel to hook reduce op.
# This is the 4th commit message:
Init mixed precision training interface (#16856 )
* Init mixed precision training interface
* Add fp16 test script
test=develop
* All initializers support float16
test=develop
* Code cleanup & add more code annotations
test=develop
* Update API spec
test=develop
* Add usage example in doc
test=develop
# This is the 5th commit message:
fix reference_count_pass,test=develop (#17060 )
test=develop
# This is the 6th commit message:
Speedup roi_perspective_transform op by caching the information of linear interpolation in forward (#17090 )
* Cache the information of linear interpolation in forward and use it in backward.
test=develop
* Fix cuda kernel.
test=develop
# This is the 7th commit message:
remove unnecessary prepare_data (#17080 )
test=develop
# This is the 8th commit message:
fix interpolate cu. test=develop (#17101 )
# This is the 9th commit message:
test=develop, double backward leaky_relu (#17067 )
backward of backward: leaky_relu
# This is the 10th commit message:
fix fuse optimizer ops (#17102 )
test=develop
# This is the 11th commit message:
truncated_gaussian_random supported in distributed training, test=develop (#17091 )
# This is the 12th commit message:
Detailed coordinate description for yolov3 loss (#17007 )
* Detailed coordinate description for yolov3 loss
test=develop
* modified api.spec
test=develop
* modified loss name
* fix api.spec
test=develop
* polish description
test=develop
* modified api.spec
test=develop
# This is the 13th commit message:
fix test_weight_decay (#17109 )
test=develop
# This is the 14th commit message:
Path flag (#17105 )
* fix python/paddle/fluid/__init__.py detecting problems
6 years ago
9ccce576d6
fix test_weight_decay ( #17109 )
...
test=develop
6 years ago
7da7881c0e
Detailed coordinate description for yolov3 loss ( #17007 )
...
* Detailed coordinate description for yolov3 loss
test=develop
* modified api.spec
test=develop
* modified loss name
* fix api.spec
test=develop
* polish description
test=develop
* modified api.spec
test=develop
6 years ago
7330cd639c
truncated_gaussian_random supported in distributed training, test=develop ( #17091 )
6 years ago
258e000be6
test=develop, double backward leaky_relu ( #17067 )
...
backward of backward: leaky_relu
6 years ago
10c487eb21
fix interpolate cu. test=develop ( #17101 )
6 years ago
55ce36e981
Speedup roi_perspective_transform op by caching the information of linear interpolation in forward ( #17090 )
...
* Cache the information of linear interpolation in forward and use it in backward.
test=develop
* Fix cuda kernel.
test=develop
6 years ago
beda78258f
Init mixed precision training interface ( #16856 )
...
* Init mixed precision training interface
* Add fp16 test script
test=develop
* All initializers support float16
test=develop
* Code cleanup & add more code annotations
test=develop
* Update API spec
test=develop
* Add usage example in doc
test=develop
6 years ago
0b07eef118
ParallelDyGraph with GPU collective mode ( #16827 )
...
implement dygraph.parallel.DataParallel to hook reduce op.
6 years ago
1a4a51db2b
Fleet unify distributed training ( #16791 )
...
* implement distributed transpiler with fleet
6 years ago
e707119a89
remove ut test_dist_word2vec in mac ci, will fix it in private, test=develop ( #17066 )
6 years ago
b8c166f632
Merge pull request #17029 from wzzju/add_graph_checkpoint
...
add checkpoint functions for graph. test=develop
6 years ago
2deac4e447
Fix the bug of test_conv2d_int8_mkldnn case which raised by improper parameter passing ( #17058 )
...
* resolve #17057
Fixed the bug that fuse_relu/fuse_residual option couldn't be passed to class TestConv2dInt8Op.
test=develop
* Fix the bug of test_conv2d_int8_mkldnn case which raised by improper parameter passing.
test=develop
6 years ago
a2be4b4d91
Add fuse momenutum ops ( #16745 )
...
* Add fuse momenutum ops
6 years ago
5e216fcf1a
fix split
...
test=develop
6 years ago
e296e0fead
fix test_parallel_executor_seresnet random fail ( #17030 )
...
test=develop
6 years ago
b3a11943c1
Merge pull request #17031 from luotao1/reduce_test_time
...
reduce unittest time by rename testcuda to has_cuda
6 years ago
c1c2633a63
Support backward of backward for Relu and add a new gradient checker by comparing theoretical and numerical Jacobian. ( #16862 )
...
* Support backward of backward and a new gradient checker
* Rename decorators.py to decorator_helper.py, since Python on Windows CI has decorators package.
1. Add ReluDoubleGradMaker when register relu_grad.
2. Add a new gradient checker by comparing theoretical and numerical Jacobian. Check double gradients by double_grad_check.
6 years ago
f188b3708e
Move gc test to each test of op ( #16999 )
...
* move gc test to op_test
test=develop
* Revert "move gc test to op_test"
This reverts commit cf15da65c38f57c91f53b3d8b3c2365d4aa86016.
* enable gc test in some ops
test=develop
6 years ago
7c370e42f9
Fix test_recurrent_op ( #17001 )
...
* fix ramdom fail
test=develop
6 years ago
9466e956a7
reduce unittest time by rename testcuda to has_cuda
...
test=develop
6 years ago
bf6b9d6d37
add checkpoint functions for graph. test=develop
6 years ago
d9991dccdd
add parallel build script to ci … ( #16901 )
...
* add parallel build script to ci test=develop
* 1. classify the test case as single card/two cards/multiple cards type
2. run test case according to the run type
6 years ago
24923f7604
fix py_reader demo ( #16997 )
...
test=develop
6 years ago
ea42e431f8
Speed unit testing. ( #16978 )
...
* Speed affine_channel_op unit testing
* Add check in tensor_py
* Fix ONLY_CPU Compiling
6 years ago
ae7a2cb8e3
resolve #16988 ( #16995 )
...
Update the filter generation mechanism that it could generate the negative parameter.
The original calling(np.random.random()) couldn't simulate the conv/relu fusion case.
test=develop
6 years ago
765c70a1b0
Unittest improve, test=develop ( #16941 )
...
* accelerate test_ir_memory_optimize_nlp, test=develop
* accelerate test_ir_memory_optimize_nlp, test=develop
6 years ago
23df084b32
resolve #16987 ( #16994 )
...
Rename the testcuda function to has_cuda, it will elimate the unnecessary testing.
test=develop
6 years ago
1202d3fc74
Refine model gpu memory ( #16993 )
...
* speedup gc and inplace softmax_with_cross_entropy_grad
test=develop
* refine models gpu mem
Merge skip vars and warning messages of mem opt
remove relu mem opt
test=develop
* follow comments
test=develop
6 years ago
af8a041bb6
reduce py_reader unittest time ( #16996 )
...
test=develop
6 years ago
45ebafc33e
Split test_calibration test to two tests ( #16977 )
...
* Split test_calibration test to two tests
test=develop
* Modify CMakeLists file
test=develop
6 years ago
3c375751f8
Support seq len equal to 0 in sequence ops ( #16935 )
...
* Support seq len equal to 0 in sequence ops
test=develop
* Add more test cases
* Fix some comments
test=develop
* Fix py3 error
test=develop
6 years ago
9d2f7d762c
fix dy-load bug, test=develop
6 years ago
94c4cd1a98
fix dy-load bug, test=develop
6 years ago
a3f17280a3
fix dy-load bug, test=develop
6 years ago
36c05d36ab
Check some shapes only in runtime ( #16919 )
...
* Check some shapes only in runtime
test=develop
* Follow review comments
test=develop
* Update API spec
6 years ago
486f7d8ed6
Restore quantization and distillation stategy before loading persistables. ( #16958 )
...
test=develop
6 years ago
cbdb8a17b1
Polish DGC code ( #16818 )
6 years ago
dbf66dd034
Merge pull request #16954 from junjun315/fix-dygraph-checkpoint
...
Fix dygraph checkpoint bug
6 years ago
aa9caa1691
Merge pull request #16951 from luotao1/reduce_ci_time
...
use multi-thread to speedup CI tests
6 years ago
aed702cea3
Merge pull request #16920 from qingqing01/test_profile
...
Fix test_profiler when the machine has many cores.
6 years ago
b596eed73a
Merge pull request #16824 from LeoZhao-Intel/mkldnn_mul
...
disable test_elementwise_mul_mkldnn_op case
6 years ago
3beed54cdd
Merge pull request #16917 from velconia/dygraph_untrack_op
...
imperative fix tracer train mode
6 years ago
a7c11979ba
fix dygraph save/load checkpoint error, test=develop
6 years ago
bc037c13c7
use multi-thread to speedup CI tests
...
test=develop
6 years ago
2b61db07d1
fix sampling id op bug ( #16909 )
...
* fix sampling id op bug, test=develop
6 years ago
b7f20ed6af
Fix unittest dataset error ( #16925 )
6 years ago
d5a7c09856
Merge pull request #16798 from phlrain/softmax_cross_support_high_rank
...
softmax cross entropy support high rank
6 years ago
b73a71d11e
Fix test_profiler when the machine has many cores
...
test=develop
6 years ago
03a7ec55cf
Merge pull request #16746 from xjqbest/dataset_merge_develop
...
move split filelist from trainer.py to fleet & fix error
6 years ago
5d45eb06f9
Merge pull request #16858 from heavengate/fix_yolo_param
...
Fix yolo param
6 years ago
97aa1838bc
Fix dygraph train mode
...
test=develop
6 years ago
102fc8596e
Merge pull request #16777 from velconia/dygraph_untrack_op
...
Imperative tracer does not hold op any more
6 years ago
1edcd73115
remove unnecessary new line
...
test = develop
resolve #16764
6 years ago
61cc842a53
disable test_elementwise_mul_mkldnn_op case
6 years ago
ccc3bd70c1
polish doc for uniform_random and multi_box_head ( #16864 )
6 years ago
0701c2db47
Merge pull request #16518 from zhoukunsheng/rsqrt
...
Rsqrt
6 years ago
bbcfa8ffb2
Merge pull request #16493 from zhoukunsheng/zeros_like
...
Zeros like
6 years ago
d722841622
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into softmax_cross_support_high_rank
6 years ago
f9223c5fa9
Logical compare ( #16513 )
...
* test=develop
update API.spec with hash, resolve conflict
* remove unused parameter
6 years ago
0114f1d65a
give the output value to keep consistant with document test=develop ( #16740 )
6 years ago
766c868199
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into softmax_cross_support_high_rank
6 years ago
485bc6a055
Merge pull request #16868 from chengduoZH/speedup_test_parallel_executor_transformer
...
Reduce the layer number of transfromer model
6 years ago
d4b5510c00
Merge pull request #16860 from junjun315/fix-utest-vgg
...
Fix bug: long vgg-utest testing time
6 years ago
072db0938b
optimize lstmp and sample_logits op, test=develop ( #16845 )
...
* optimize lstmp and sample_logits op, test=develop
* update op_use_default_grad_op_maker.spec, test=develop
* delete useless file,test=develop
* append 0 to dim variable to avoid memory reusage, test=develop
6 years ago
Shuai Yuan
wopeizl
Zeng Jinle
Krzysztof Binias
Hongyu Liu
guru4elephant
Guo Sheng
chengduo
flame
Kaipeng Deng
mozga-intel
pkpk
Yibing Liu
Jiabin Yang
pkpk
Qiao Longfei
zhang wenhui
lijianshe02
Sevin F. Varoglu
Yiqun Liu
guomingz
Zhen Wang
Tao Luo
zhaoyuchen2018
Zhaolong Xing
liuwei1031
lvmengsi
jiaqi
Yan Xu
Jie Fang
Yan Xu
baojun
JesseyXujin
jerrywgz
zhoukunsheng
qingqing01
SunGaofeng
wopeizl
xiaoting
tangwei12
lujun
Huihuang Zheng
gongweibao
tensor-tang
ruri
tianshuo78520a
Lfc1993
Jiancheng Li
whs
tink2123
Wang, Chuanqi
gongweibao
minqiyang
Qiyang Min
Leo Zhao
phlrain