9f926eb720
Layernorm opt ( #29522 )
...
* layernorm fw opt
* layernorm bw opt
* fix typo, test=develop
* remove const dim3 for windows CI compatibility
* merge develop
Co-authored-by: zlsh80826 <zlsh80826@gmail.com>
5 years ago
b781953ef5
[oneDNN] Fix flags use test for #29080 , assert condition more general ( #29493 )
...
* Flags assert condition more general, print output if pattern not found
* removed test_flags_use_mkldnn form skip list regarding #29080 descr
5 years ago
ae3f7a7100
add ps table ( #29463 )
...
* add ps table
Change-Id: I468a04bd071d21ff52654926fcf4d5f3da19e178
5 years ago
36ec9456cf
Make PADDLE_ROOT as an environment variable
5 years ago
d8391a1983
fix error message of gather nd ( #29521 )
5 years ago
5ac71b36fb
Remove tensor copy in the update_loss_scaling op. ( #29426 )
...
* remove tensor copy in the update_loss_scaling op
* not use thrust.
* fix some cuda memory access error.
5 years ago
e74e1a226c
support deepcopy for Layer/Tensor/Paramerbase ( #29387 )
...
* support deepcopy for Layer/Tensor/Paramerbase
* fix some code
5 years ago
87e75a77c2
Add tangent operator ( #29207 )
...
As the title
5 years ago
95e334810a
Softmax vectorization ( #29404 )
...
* vec softmax fw
* vec softmax bw
* add a message argument for compiler compatibility
5 years ago
a136c9cdb8
fix increamental coverage script bug, WITH_INCREMENTAL_COVERAGE to DWITH_INCREMENTAL_COVERAGE, test=develop ( #29509 )
5 years ago
966aa0e387
Fix test_mobile_net random failed on windows GPU( #29480 )
5 years ago
2ef9e0e23c
Rebuild group automatically in dynamic graph distributed ( #29255 )
...
* add tensor_indices in AssignGroupBySize
* add rebuild group in reducer
5 years ago
3a0558339d
support mobilenet for kunlun ( #29458 )
5 years ago
a1909affc6
Fix Unit Test: Add Sleep Time for CUDA Retry ( #29442 )
...
Add Sleep Time for CUDA Retry, which is similar to our GPU retry logic. This is a try to avoid init GPU allocation random failure in unit test.
5 years ago
e5e522493d
make gelu fp16 computing more robust ( #29484 )
5 years ago
8094ac686e
Print ccache/clcache hit rate ( #29341 )
...
* test ccache hit statistics, test=develop
* test ccache hit statistics, test=develop
* add cache hit statistics, test=develop
* fix no percent symbol erro on windows, test=develop
* remove switch, test=develop
5 years ago
560b432349
Revert "improve elementwise_add_grad perf ( #29277 )" ( #29464 )
...
This reverts commit befd6d5338
5 years ago
57a4f16d9e
added internal and external reorders to profiler ( #29443 )
...
* added external reorder to profiler
* added external and internal reorders to profiler
* added internal and external reorder to profiler
* added formatting to int/ext reorder commit
* removed unnecessary comment
5 years ago
2480bdef6c
change hard_swish from plugin to layer ( #29177 )
...
* change hard_swish from plugin to layer
* add ut when threshold != scale
5 years ago
ecca6585cd
1. fix elementwise ops'bug 2. fix softmax_with_cross_entropy_op 3. add biliner_interp_op ( #29448 )
...
Co-authored-by: root <root@bjhw-sys-rpm0223.bjhw.baidu.com>
5 years ago
03b42d9fa7
fix unittest on windows, test=develop ( #29365 )
5 years ago
a5fcc4b545
update reduce_sum op on xpu ( #29367 )
...
* update reduce_sum op on xpu
* update reduce_sum op on xpu
* support running on xpu
5 years ago
c7cada8571
Fix gru performace decline in 1.8.5 ( #29455 )
5 years ago
6296f4ed09
revert cast eigen kernel ( #29427 )
5 years ago
a040c055a5
fix layer_norm accuracy ( #29434 )
5 years ago
24ba9ed436
fix that parameters'grad has grad var ( #29408 )
5 years ago
4e19ce1df5
refine reshape grad and double grad kernel, use tensor copy async ( #29128 )
5 years ago
225a9c4ed8
Fix unittest ( #29412 )
...
* fix tensorrt unittest precision error
* fix unittest precision error. test_trt_subgraph_pass && test_trt_dynamic_shape_transformer_prune
5 years ago
f860de4af7
support clip op trt converter ( #29411 )
5 years ago
1dd7b97b66
fix rnn_op bug in cudnn_version>= 8 ( #29406 )
5 years ago
671555ed32
Compiling operator libraries with Unity build ( #29130 )
...
* Compiling operator libraries with Unity Build on Windows CPU.
* Compiling operator libraries with Unity Build on Windows GPU, no_test, test=windows_ci
* Add option in windows ci script, no_test, test=windows_ci
* Optimize parallel compiling, test=develop
* remove limit of parallel compile and skip some ops in UB, test=develop
* remove changes of header file, test=develop
* remove changes of header file, test=develop
* fix test_eye_op unittest failed, test=develop
* Compiling operator libraries with Unity Build on Linux, test=develop
* set default WITH_UNITY_BUILD=OFF, test=develop
* Move unity build rules into a single file and add comment, test=develop
* optimize parallel compilation, test=develop
* fix undefined reference error on coverage ci, test=develop
5 years ago
5c9bd0bf7c
print whether has build cache ( #29035 )
5 years ago
a623ce044f
Use different name_scope for different conv type, test=develop ( #29355 )
5 years ago
7c508d8668
update unbind norm add CUDAPlace api doc information ( #29322 )
...
* enhance array_to_lod_tensor_op lod_tensor_to_array_op errors information. test=develop
* fix format. test=develop
* format fix. test=develop
* add lod_rank_table. test=develop
* fix format. test=develop
* fix doc info. test=develop
* fix np error
* add unbind dygraph api. test=develop
* fix unbind doc.test=develop
5 years ago
879e913b6d
Make transpose, trace, kron, reshape, sum op support complex type ( #29321 )
...
* add complex64 and complex128 type; add +-*/@ and slice opreator for complex types
* add test cases for complex elementwise, matmul and getitem unittest
* add test cases for complex types
* add test cases for complex matmul unittest
* kron, reshape, transpose support complex types
* sum and trace op support complex types
* add test case of sum and trace op
* fix the bug of imag part of complex not initialized
* format file
* format code style
* kron support type promotion; modify test cases
5 years ago
074065e5de
fix expand/uniform_random && concat/transpose to new api on xpu ( #29280 )
...
* fix expand && concat/transpose to new api
* update uniform_random_op
* update xpu_header
5 years ago
1decf4ada6
update, test=develop ( #29331 )
5 years ago
74bf3bed36
support global pooling for kunlun ( #29293 )
...
* test=kunlun
5 years ago
b10ecd9d3a
[inplace] Add ShareHolderWith for class Variable and SharePlaceholderWith in VarBase.detach() to share the same Tensor/SelectedRows ( #29267 )
5 years ago
9ad800ebb2
Support type promote for basic math ops (quantum required) ( #29265 )
...
* basic impl of type promote
* add comment & another testcase
* fix complex bugs & support python op promote type
* fix failed unittests & polish code
* add unittest for coverage
* change to only promote complex type
* polish code details
* polish several comments
5 years ago
8358791607
fix gpu outofrange ( #29238 )
...
* fix gpu emb out of range
Change-Id: I5794ac73bd634d5ea069a6fbbd914274b6d6b7bf
* fix doc
Change-Id: I5a3350b2930a9ab2f52116c192b087307faf8fdf
5 years ago
b58cfff89d
use has_grad instead of train_mode ( #29309 )
...
* use has_grad instead of train_mode
* add vlog for debug
* fix ut
* fix ut
5 years ago
befd6d5338
improve elementwise_add_grad perf ( #29277 )
...
* improve performance of elementwise_sum_grad
5 years ago
ebf689197d
fix tensorrt output shape error ( #29308 )
...
* fix tensorrt output shape error
* fix unittest tensorrt_engine_op_test
* fix code style for unitest
5 years ago
67c700b479
[Dy2Stat] Add cache for Executor and Context in run_program_op ( #28421 )
5 years ago
696dc4bb13
fix the warning of reducer ( #29323 )
5 years ago
c4be80f402
polish the code of cumsum and remove some unused code ( #29303 )
5 years ago
c00af94435
fix matmulv2 for windows ( #29302 )
5 years ago
3765da98c7
add coverage incremental switch, test=develop ( #29290 )
5 years ago
d68af02c04
fix analysis_config bug. ( #29304 )
5 years ago
0fb18bc214
enforce the matmul_v2 error message ( #29297 )
5 years ago
9b59a589b1
Remove some useless log. ( #29300 )
5 years ago
13a22a3752
fix shape of tile_grad op ( #29289 )
5 years ago
be3777a50a
Add pure fp16 training with master weights. ( #27712 )
...
* add the weight decay func for the momentum op
* Add the multi_precision function in Momentum Optimizer.
* Make sure that the initial value of master weights are same with the fp16 weights.
* add static loss scaling.
* add the rescale_grad function in the pure fp16 training.
* use the original momentum updating method.
* Polish some codes, such as variable names.
* add docstring for apis.
* update the var creation details of _create_master_weight.
* not modify codes about imperative momentum updating.
* Fix the error of test_dist_sparse_tensor_load_momentum UT.
* add unit test for multi precision fp16 training.
* add more unit tests for CI.
* Use lower threshold values for allclose comparing in test_multi_precision_fp16_train UT.
* For CI Coverage Checking.
5 years ago
6673fb0565
change import math.h to cmath ( #29260 )
5 years ago
7584bb5096
Layer norm fp16 ( #29169 )
...
* add fp16 for layer_norm op
* revert layernorm api
* fix forward
* fix forward
* fix backward for layernorm with fp16
* fix unit test for layernorm with fp16
* fix with_mkldnn compile error for layernorm with fp16
* 1. revert to PADDLE_ENFORCE_NOT_NULL, 2. change static_cast<float> to static_cast<U>
* fix with_mkldnn compile error for layernorm with fp16
* fix with_mkldnn compile error for layernorm with fp16
Co-authored-by: zhiqiu <chenqiuliang@baidu.com>
5 years ago
c59b4f28a2
fix cmake error when WITH_GPU=ON and WITH_TENSORRT=ON && WITH_MKL=OFF ( #29275 )
5 years ago
fc80d2e09c
add compile option WITH_TENSORRT ( #29208 )
...
* add compile option WITH_TENSORRT
* add WITH_TENSORRT to ci paddle_buils.sh
* add WITH_TENSORRT to paddle_build.sh
* change FATAL to WARNING when TensorRT is not found and WITN_TENSORRT=ON, just to pass ci-py3 temporarily
5 years ago
116305ea4b
Improve performance of elementwise_add grad op ( #29187 )
...
* pass stop_gradient for cast op
* improve performance of elementwise_add grad
* use tensor copy async
* dygraph branch
* fix dygraph branch
* add ut
5 years ago
07c67d5a8b
add deformable_conv op on xpu ( #29234 )
...
* rebase develop
* update deformable_conv op on xpu
* update deformable_conv op on xpu
5 years ago
1de32f823d
Hot fix complle failed in gcc4.8 caused by complex impl ( #29254 )
...
* hot fix complle failed in gcc4.8
* fix failed unittest
5 years ago
642abe2a48
Fix a bug when running on an operating system without "bash." ( #29131 )
...
* Fix a bug when running on an operating system without "bash."
* add execution condition
* for ci-coverage
5 years ago
46b73e6cd9
Change the api of DataParallel and Fleet ( #29224 )
5 years ago
64f29fbb70
update kunlun conv2d/softmax/elementwise implemetation ( #29229 )
...
* update conv2d & softmax to new xpu api
* test=kunlun
* remove useless comments
* test=kunlun
* remote softmax xpu op
* test=kunlun
* update kunlun softmax
* test=kunlun
* update xpu unitest
* test=kunlun
* fix elementwise_grad bug for kunlun
*test=kunlun
5 years ago
8f45d14263
add complex64 and complex128 type; add +-*/@ and slice opreator for c… ( #29199 )
...
* add complex64 and complex128 type; add +-*/@ and slice opreator for complex types
* add test cases for complex elementwise, matmul and getitem unittest
* add test cases for complex types
* add test cases for complex matmul unittest
5 years ago
c0a991c874
accumulate gradient for leaf tensor with previous graph and expose leaf tensor concept ( #28429 )
...
* The leaf tensor concept is exposed and the gradient accumulation of leaf tensor
* The leaf tensor concept is exposed and the gradient accumulation of leaf tensor
* fix coverage
* fix api doc
* fix CI unittest
* fix CI unittest
* fix unitest
* empty tensor does’t need inner_var_
* fix some error message
5 years ago
74c43ac638
fix lite unit test. ( #29233 )
5 years ago
4096ff94dc
Small optimizations for conv2d kernel subroutines. ( #29188 )
...
- Make sure that oneDNN memory descriptors are created only once at
first iteration.
5 years ago
5c61eeef61
Enable all image classification models ( #29155 )
5 years ago
4fec182d24
[Lite-Subgraph] Fix compile error for lite subgraph. ( #29146 )
5 years ago
b5c6342336
Update ps gpu ( #29209 )
...
* fix paramete prefetch & device guard
Co-authored-by: MrChengmo <cmchengmo@163.com>
Co-authored-by: chengmo <chengmo@baidu.com>
5 years ago
865a45984f
Check whether there is any inplace operation affecting gradient calculation. ( #27901 )
...
* Add a class TensorInplaceVersion to count the inplace version and put it in framework::Tensor instead of Allocation or Variable.
* Add a new attribute `_inplace_version` for VarBase.
* Raise exception if an inplace operation can result in incorrect gradient computation.
* Add a new interface _bump_inplace_version() for VarBase to bump the version whenever the Tensor is modified through an inplace operation.
* For api assign, call _bump_inplace_version() when it's an inplace operation inn dynamic mode.
* Use original var_wrapper if the inplace_version is not changed.
* Replace SnapshotVarWrapperList with SnapshotVarWrapper to optimize performane.
5 years ago
4056c4f11c
Add unittest in musl build ( #29099 )
...
* add musl docker build script
* rm space test=document_fix
* fix some docs and types errors test=document_fix
* move install of python requirement to docker build
* add copyright to docker file.
* add extr opts
* format docs
* add ut test add pip cache
* add more args description in readme
* add stack backtrace in ctest
* fix readme bugs
5 years ago
03d4665f44
prefetch optimize ( #29095 )
...
* test=develop, optimize async prefetch
5 years ago
0c2a51d240
optimizer amp, all use fp16 communication, overlap last comm and compute ( #28957 )
5 years ago
0b032faeee
Polish unittests details and execution conditions to adapt to MUSL ( #29044 )
...
* fix failed tests in yingchun gived list
* add unittests into static_mode_white_list
* add enable static
* fix dist unittest
* skip test_sigmoid_focal_loss_op & add gym
* revert no need skip unittests
* remove gym
5 years ago
92817f8005
test=develop, rm pathlib ( #28658 )
...
* test=develop, rm pathlib
5 years ago
4fd4095d1b
Add quantization of multi_gru op and tests ( #28615 )
5 years ago
bc6033f86b
fix gru gcc7.4 bug for the gru compile
...
fix gru gcc7.4 bug for the gru compile
5 years ago
0239f79695
Generate code coverage reports only for incremental files ( #28508 )
...
* Generate code coverage reports only for incremental files, test=develop
* Generate code coverage reports only for incremental files, test=develop
* Generate code coverage reports only for incremental files, test=develop
* test for diff python file, test=develop
* fix no python diff report, test=develop
* add cc test file, test=develop
* fix bug in generic.cmake, test=develop
* for debug no cc report, test=develp
* modify compire branch form test_pr to test, test=develop
* fix bug, test=develop
* test for h file changed, test=develop
* debug for redefinition of argument optimize error, test=develop
* close -o3 for test, test=develop
* remove -o3 for test, test=develop
* remove coverage option for nvcc, test=develop
* use CMAKE_CXX_FLAGS open coverage option when header file changed, test=develop
* reopen -o3, test=develop
* remove debug code, test=develop
* remove unused code, test=develop
5 years ago
b818429ae7
optimize cumsum OP ( #29193 )
5 years ago
e2d01eb650
Support dynamic graph distributed ( #28997 )
...
* add reducer
* refine envent for memorycopy
* add concat&split for allreduce
* apply concat & split for fuse tensor
* fix nccl dep
* fix the untest, compile problem and ddp initialize problem
* fix untest for mac & add some comments & solve the repeated param in sublayers
* fix untest for windows & fix document
5 years ago
7e5e9934fe
update expand as op to use the shape of the target tensor instead of the target tensor itself. ( #29020 )
...
* update, test=develop
5 years ago
7c8ac064c8
Delete prettytable in condabuild ( #29145 )
...
* update conda_build script with removing opencv
* modified filepath
* modified some content
* Delete Commented-Out Code
* delete prettytable in conda_build
Co-authored-by: XieYunshen <1084314248@qq.com>
5 years ago
e668cb07fb
fix CUDA 11 error on windows ( #29101 )
5 years ago
085260f3de
Add eigen gru and fix the dropout bug in the rnn
...
Add eigen gru and fix the dropout bug in the rnn
5 years ago
545df287fc
add user_define_dump ( #28596 )
5 years ago
71815637cc
Move gym into unittest/requirements.txt ( #29149 )
5 years ago
bc902044a4
Fixes mkldnn dygraph learning rate scheduler crashes ( #28988 )
5 years ago
b9e76a0103
detect tensorRT plugin fp16 in runtime ( #27933 )
...
* remove -DSUPPORTS_CUDA_FP16 in cuda.cmake
* comile with cuda9
* add some unittest
* notest;test=coverage
* add unittest for trt plugin swish && split
* update ernie unittest
* fix some error message
* remove repeated judgement of CUDA version in mbEltwiseLayerNormOpConverter
* fix comile errror when CUDA_ARCH_NAME < Pascal"
* fix comile error
* update unittest timeout
* compile with cuda9
* update error msg
* fix code style
* add some comments
* add define IF_CUDA_ARCH_SUPPORT_FP16
* rename IF_CUDA_ARCH_SUPPORT_FP16 to CUDA_ARCH_FP16_SUPPORTED
5 years ago
fd3fcb051a
fix typo of flag name ( #29154 )
5 years ago
da71173bc9
Fix ops doc for some ops
...
Fix ops doc for some ops
5 years ago
770395cb93
Split train_mode and has_grad for tracer ( #29064 )
...
* split train_mode and has_grad
* fix format
* fix ci problems
* fix sample code
5 years ago
7ae3cb554a
Polish CUDA Information stdout ( #29109 )
5 years ago
7a15e64034
Support precision test for new ut
5 years ago
173c22aec2
optimize fast graph executor ( #28962 )
5 years ago
562ded1041
fix unittest trt_dynamic_shape_transformer_prune_test error ( #29122 )
5 years ago
db41258501
add API serialize_program, serialize_persistables, save_to_file, deserialize_program, deserialize_persistables, load_from_file. ( #29034 )
5 years ago
b0d1ac161e
Add bf16 pool2d and unify bf16 unit tests ( #29039 )
...
* Add bf16 pool2d and unify bf16 unit tests
* Add change default ops test
5 years ago
fddea67445
Fix cpu_bfloat16_pass ( #28730 )
...
* Fix cpu_bfloat16_pass
* Add output_format
* Fix incorrect SetOutput
* Change fromating
5 years ago
2fd16cf6fc
fix win ci failure, test=develop ( #29089 )
...
* fix win ci failure, test=develop
* add ci test, test=develop
5 years ago
fea0e294ee
Hide the C++ stack by default and add hints ( #29042 )
...
* default not show cpp statck & add hint
* fix failed unittest
* fix failed unittests
5 years ago
b1274ac3d6
set show cpp stack by default, test=document_fix ( #29102 )
5 years ago
582c0a0468
add uint8 for reshape op ( #28996 )
...
add uint8 for reshape operator
5 years ago
8ca0a8a859
fix tensor detach to zero copy ( #27921 )
...
* fix tensor detach to zero copy
* fix tensor detach to zero copy
5 years ago
8af0d85ea4
fix unittest failed on windows GPU ( #29072 )
5 years ago
a5aa4dc7a9
add xpu elementwise ops ( #29031 )
5 years ago
b04c78ef5e
Update pow ( #29000 )
...
Simple code clean up
5 years ago
b2c8a00745
remove eigen threadpool for the speed up
...
remove eigen threadpool for the speed up
5 years ago
7b5a8e46de
Add multi_gru_fuse_pass and tests ( #28601 )
...
* Add multi_gru_fuse_pass and tests
* fix date
* cleaned up headers
5 years ago
c91bb084f4
Add op benchmark ci pipeline in Paddle repo ( #28692 )
5 years ago
5e26a15484
Open GPU unitest on windows ( #29003 )
...
* open unittests on windows
* open GPU unittest on windows
5 years ago
3815d7aa40
Upgrade string literals to raw string ( #28989 )
...
* upgrade comment string to raw string
* fix string in
* fix string with ' '
* revert update on comments
* upgrade only necessary
* fix sample code checker
* fix comments with '''
5 years ago
767d0ba267
update, test=develop ( #28700 )
5 years ago
991345b368
Add multi_gru_seq_fuse_pass and tests ( #28604 )
...
* Add multi_gru_seq_fuse_pass and tests
* fix date
* removed unused functions
5 years ago
fbf9564f6b
【paddle.distributed.fleet】Optimize ParameterServer's Async Mode ( #28442 )
...
* test=develop, optimize global_step
5 years ago
f77a78cdee
enable pipeline to run with Executor.run() ( #28373 )
...
* update, test=develop
5 years ago
0073f9bdb0
support ps-gpu ( #28752 )
...
* ps gpu transpile
* ps gpu
* remove op
* gps trainer
* local ps
* add macro
* HeterBox
* def cuda
* tab
* code style
* style
Co-authored-by: Thunderbrook <a754913769#163.com>
5 years ago
768dab441e
polish two api doc detail, test=document_fix ( #28971 )
5 years ago
8ff3550658
refactor momentum op to combine weight ( #27414 )
...
* refactor momentum op to combine weight_decay (scale op and sum op)
5 years ago
bd1d6d3b30
extends oneDNN caching keys so caching objects are unique to executor/predictor ( #28758 )
5 years ago
3d0ff8eebc
optimize musl docker build script ( #28974 )
...
* add musl docker build script
* rm space test=document_fix
* fix some docs and types errors test=document_fix
* move install of python requirement to docker build
* add copyright to docker file.
* add extr opts
* format docs
5 years ago
994673bf4f
change avg pooling and global pooling to trt layer in dynamic shape mode ( #28702 )
...
* change avg pooling and global pooling to trt layer
* add support for static shape global pooling
* modify trt errmsg
5 years ago
71c1cd1408
fix truncated_gaussian seed ( #28777 )
5 years ago
de528981e5
fix paddlepredictor build error. test=develop ( #28792 )
5 years ago
a22ea652cf
fix trt delete_pass bug. ( #28763 )
5 years ago
1dad8ceaab
Fix gpu memory allocation bug. ( #28703 )
5 years ago
b969c32ab1
fix occupied 0 device memory bug ( #28771 )
5 years ago
1a532d5133
add uint8 support for squeeze operator ( #28734 )
...
Adding uint8 support for squeeze operator.
5 years ago
8b853b3030
fix the number of perf algo for conv cudnn in exhaustive mode ( #28694 )
5 years ago
8c0ea4bffe
Add bf16 matmul, fc, elementwise add and mul ( #28729 )
...
* Add bf16 matmul, fc, elementwise add and mul
* Correct unit test
5 years ago
efc3b182f0
a fix for the fc_lstm_fuse_pass ( #28709 )
5 years ago
3b0dd5f620
fix bug that to_tensor not support paddle.Place ( #28717 )
5 years ago
08b62f4902
fix shuffle batch op shuffle ( #28533 )
5 years ago
d3d1a6b6e0
add kunlun kernel: slice, slice_grad, top_k, cast. *test=kunlun ( #28542 )
...
* 1.add xpu slice op 2. add xpu top_k op 3.modify xpu cast to new api
* 1.add xpu slice op 2. add xpu top_k op 3.modify xpu cast to new api
5 years ago
9362d85e0e
Add LSTM, Simple RNN and GRU CPU kernel ( #28577 )
...
* add lstm, simple rnn op kernel
* fix the test_lstm for the rnn op
* change func name
* fix forward postprocess bug
* add gru forward, backward code
* remove unittest.skipIf; use a big rnn op instead of combination op
* fix input doesn't have gradient bug
* add eigen lstm forward, backward
Co-authored-by: wawltor <fangzeyang0904@hotmail.com>
5 years ago
30ef3815b3
adjust kunlun header file ( #28536 )
...
* adjust kunlun header file
*test=kunlun
* update kunlun unittest
*test=kunlun
* update xpu unitest
* test = kunlun
* update xpu unittest
* test=kunlun
* update xpu unitest
* test=kunlun
5 years ago
dab4920568
improve performance of cast op ( #28727 )
5 years ago
3a88acd2ee
open unittests on windows ( #28750 )
5 years ago
03f46e3526
fix truncated_gaussian op cuda seed setting ( #28678 )
5 years ago
04cefeacc5
Disable windows gpu static lib. ( #28741 )
5 years ago
04bcc13fac
Add multi_gru op and tests ( #28591 )
...
* Add multi_gru op and tests
* removed redundant disable_dygraph()
5 years ago
5aec7dbeb0
use forward declarations for framework.pb.h ( #28494 )
...
* use forward declarations for framework.pb.h, test=develop
* use forward declarations for framework.pb.h, test=develop
5 years ago
f1074e3b19
hide the token output to safely ( #28716 )
5 years ago
32b90b1c2d
add log10 ( #28576 )
...
Add new operator log10
5 years ago
3d09929b1f
Add check for non-dispensable input ( #28666 )
...
* Add check for non-dispensable input
* fix typo
5 years ago
7eeb99fe02
Add basic hook classes for dygraph & implement reduce hook ( #28584 )
...
* add base hook classes and reduce hook impl
* fix constructor typo
* polish comment format
* refactor baisc hook class design
* polish design details
5 years ago
858ffa0c8b
Fix the dropout setting when not initialized in rnn_op. ( #28561 )
...
test=develop
5 years ago
6d8d3d4c22
[oneDNN] Layer norm bf16 kernel ( #28619 )
5 years ago
80d2024644
bug fix, test=develop ( #28674 )
5 years ago
bf143652ac
fix lstm OP compile error on windows ( #28667 )
...
* add unittest and check unittest for windows
* fix lstm OP compile error on windows
5 years ago
57dab959ca
add datanorm op new scale_w register ( #28657 )
...
Co-authored-by: yaoxuefeng6 <yaoxuefeng@baidu.com>
5 years ago
65aac81191
Fix fake_quant error when cout > 1024, test=develop ( #28603 )
5 years ago
b2f7ab6636
bug fix, test=develop ( #28648 )
5 years ago
8f2656ef5c
fix the gradient bug for the topk v2
...
fix the gradient bug for the topk v2
5 years ago
a972c33fd7
refine gather OP performance for dynamic mode ( #28587 )
5 years ago
2cb71c0cde
Add checkpoint to quantize ( #28612 )
...
* Add checkpoint to quantize
* Change bfloat16 option
5 years ago
804271cff9
Op version python mkldnn_inplace test ( #28354 )
...
* add mkldnn inplace op version test
* update mkldnn_inplace fuse pass
* update the inplace test
5 years ago
b889a0cee2
add gaussian_random op_version ( #28602 )
5 years ago
cf2c42a937
fix exec nightly error on mac ( #28567 )
5 years ago
110febdc54
Fix gradients with ignore_idx in softmax_with_cross_entropy ( #28622 )
...
* Fix gradients with ignore_idx in softmax_with_cross_entropy.
test=develop
* Fix gradients with ignore_idx in softmax_with_cross_entropy on cpu.
Remove softmax_with_cross_entropy from op_threshold_white_list.
test=develop
* Fix test_softmax_cross_entropy_op.py.
test=develop
5 years ago
8b97bb2e1f
Update cmake for arm ft and fix a bug for Predictor dtor. ( #28586 )
5 years ago
f962bd3432
Fix cudnn workspace limit in cudnn-8 ( #28611 )
5 years ago
90805e2df7
Register op_version for new attribute use_addto ( #28463 )
...
* register op_version for addto
* upgrade pass capability
* change eq to le
* change eq to le
* fix merge
5 years ago
a24d186814
fix nccl init failed in parallel dygraph mode ( #28497 )
5 years ago
93c39779b4
open a part of GPU unittest for windows ( #28378 )
...
* open a part of GPU unittest for windows
* open a part of GPU unittest for windows
5 years ago
ed9dd7c9f0
add send and recv ops ( #28590 )
...
* update, test=develop
5 years ago
a829357e4d
register the op version for some ops
...
register the op version for some ops
5 years ago
bf6e7cba7a
updata 2.0 API english doc ( #28525 )
...
* make Numpy version is below 1.19.3
* fix 2.0 doc
5 years ago
7b1619e69b
disable test_trt_dynamic_shape_transformer_prune,test=document_fix ( #28588 )
5 years ago
849467b5aa
fix user set CUDA_VISIBLE_DEVICES start/end with quotation marks ( #28547 )
5 years ago
8699f38d08
裁剪transformer模型trt支持；修复tensorRT不支持DeletePass的bug ( #28517 )
...
* skip_layernorm_op done
* add unittest
* slice op convertor support trt < 6
* skip_layernorm only work in ernie
5 years ago
08d2413142
add log2 operator ( #28319 )
...
As the title
5 years ago
0fc181dbd0
[Fix bug] If the pass name is not found, IsCompatible should return false ( #28475 )
5 years ago
1bf4836580
[Inference] Add TryShrinkMemory interface. ( #28409 )
5 years ago
c52fe48f6f
fix the GetKernelTypeForVar of input for fluid.gather ( #28534 )
5 years ago
d7cfee9b31
Checkout point add ( #28488 )
...
* upgrade pass capability
5 years ago
98dc11bb6a
add monitoring for executive ut at night ( #28377 )
...
* add monitoring for executive ut at night
* fix some error for paddle_build.bat
* fix some error
* fix some error in windows
* fix some error on windows
5 years ago
75196cda40
Paddle-TRT int8 support mul op channelwise quant ( #28422 )
...
* paddle-trt support mul channelwise quant
* add support for depthwise_conv2d
* add errmsg for unsupported op type
5 years ago
47cbf61dd4
fix softmax unittest float16 random error ( #28480 )
5 years ago
53e9aa948d
remove diff with develop ( #28504 )
5 years ago
369605be1d
fix cmake error when execute build_inference_lib ( #28503 )
5 years ago
645e999afc
fix api_impl test. ( #28483 )
5 years ago
1e698c600e
fix cmake error when setting ut timeout properity ( #28492 )
5 years ago
e14ed71cc2
refine the performance of gather Op ( #28458 )
5 years ago
e29ab5eacb
clear clcache cache file and reopen clcache ( #28384 )
...
* clear clcache cache file and reopen clcache, test=develop
* reopen clcache, test=develop
5 years ago
ba0756325a
exec ut no more than 15s 1 ( #28439 )
...
* disable ut test_parallel_executor_fetch_isolated_var,test=document_fix
* test for limiting ut exec time as 15S
* fix an error caused by cannot find ut
* fix some error
* can not find test_transformer
* fix error caused by ut not run in windows
* fix error caused by Compiler Options
* fix error caused by setting timeout value as 15 in python/paddle/tests/CMakeLists.txt
* setting timeout value to 120s for old ut
* add the timeout value setting
* fix error caused by ut only run in coverage_ci
* add analyzer_transformer_profile_tester
* fix some error
* fix some error
* fix error with inference option
* fix error with inference option setting as ON_INFER
* add some ut to set timeout
* modified some option
* fix error
* fix some timeout error
* fix error
* fix error
* fix timeout for test_analyzer_bfloat16_resnet50
* fix error
* setting timeout properity for some ut
* first pr for new ut timeout as 15S
5 years ago
155b4f9b6c
Remove selected rows all reduce over height check ( #28460 )
...
* remove slelected rows all reduce over height check
* polish unittest
5 years ago
fad4744aa4
fix crash in adam in xpu, *test=kunlun ( #28433 )
5 years ago
6bba8e57b1
fix batch_norm_xpu bug & remove xpusimulator dependence ( #28430 )
...
*test=kunlun
5 years ago
ced5c40c41
Update memory release interface. ( #28456 )
5 years ago
7821759d48
Add bfloat16 softmax and gelu ( #28394 )
...
* Add bfloat16 softmax and gelu
* Add pass attr bfloat16_enabled_op_types
* Changes from review
5 years ago
ba0fe0a812
revert the modified shell script ( #28453 )
5 years ago
c42e656179
Add retry for dygraph parallel socket bind ( #28404 )
...
* add retry for dygraph parallel socket bind
* change to loop always
* fix writing error
5 years ago
c41fd033e5
check op_version_registry in CI test, test=develop ( #28402 )
5 years ago
ca41541472
[oneDNN]Sum bf16 kernel ( #28382 )
...
* - Added sum bf16 oneDNN
test=develop
* - Fix to UT of sum bf16
test=develop
5 years ago
23439b1688
show cpp stack when catch signal ( #28415 )
5 years ago
44a476c2ab
support cuda pinned place ( #28416 )
5 years ago
12b9587be5
Add conv_bias pass version python test ( #28278 )
...
* add conv_bias pass version test
* update according to reviews
5 years ago
05114693cf
[Inference] Memory modification for ShrinkMemory. ( #28355 )
5 years ago
Leo Chen
arlesniak
tangwei12
chalsliu
ShenLiang
Zhen Wang
Zhou Wei
joejiong
zlsh80826
wanghuancoder
Aurelius84
procr
Huihuang Zheng
LoveAn
Zhang Ting
jakpiase
Pei Yang
taixiurong
TTerror
Jack Zhou
Shang Zhizhou
cc
yongqiangma
chentianyu03
lilong12
QingshuChen
liym27
Chen Weihang
wangchaochaohu
Wilber
Wojciech Uss
furnace
GeminiCarrie
Adam Osewski
joanna.wozna.intel
123malin
chen zhiyu
WangXi
pangyoki
yaoxuefeng
Noel
Shibo Tao
Qi Li
wawltor
Thunderbrook
Jacek Czaja
HappyAngel
gongweibao
iducn
Guo Sheng
石晓伟
lidanqing
YUNSHEN XIE
danleifeng
Zhong Hui
zhupengyang