91b2851cdc
enable pyreader use pin memory ( #14066 )
...
* enable pyreader use pin memory
* add py reader pin memory test test=develop
6 years ago
0b29078201
Merge branch 'develop' into grid_sampler
6 years ago
0c319e0b35
Add affine grid generator op ( #12238 )
...
* Add affine grid generator.
* fix ffine grid.
* Add unitest.
* Add CPU kernel and fix unitest.
* Fix CPU kernel.
* Refine code.
test=develop
* Fix python api.
test=develop
* Update python api.
test=develop
* Fix comment.
test=develop
* Rename affine_grid_generator to affine_grid and enhence unitest.
test=develop
* Fix unitest.
test=develop
6 years ago
cf1944af2a
test=develop
6 years ago
d325e668b8
[1.1] Load vars on PSERVER ( #14037 )
...
* fix dim0 in _load_slice_up_vars
* fix dim0 in _load_slice_up_vars, fix innershape in delete_var_op
* Revert "fix lookuptable in reduce strategy"
This reverts commit 0e722c5
6 years ago
e99da0b583
api change: create_variable_for_type_inference. test=develop
6 years ago
2eaa291e91
Merge pull request #14197 from luotao1/remove_with_fast_bundle_test
...
remove unused WITH_FAST_BUNDLE_TEST option
6 years ago
f76fee644c
fix graph pattern detector ( #14186 )
6 years ago
fe8f178582
fix word2vec related inference unit-tests ( #14203 )
6 years ago
e1742050ea
fix merge lod_tensor bug ( #14199 )
...
test=develop
6 years ago
0a180584e6
clean cmake. test=develop
6 years ago
85bcb286f5
refine vmul jitcode
...
test=develop
6 years ago
a764e900a5
Merge remote-tracking branch 'ups/develop' into fea/jit/gen
...
test=develop
6 years ago
a3377f7b0a
refine jitcode and add vmul jitcode implementation
6 years ago
1ace55c8ee
merge develop branch
6 years ago
9da7b33515
details
6 years ago
df4a3544aa
nearest neighbor interp add cuda kernel. test=develop
6 years ago
913b569903
Merge pull request #14151 from panyx0718/fix
...
add a small test to verify tensor type
6 years ago
da8ee1fbaa
fix API.spec not add defaults. test=develop
6 years ago
2ccf77d1c1
Refine GetTensorFromVar ( #14160 )
...
* fix GetTensorFromVar
test=release/1.1
* refine GetTensorFromVar
test=develop
6 years ago
5ac575cf62
remove unused WITH_FAST_BUNDLE_TEST option
...
test=develop
6 years ago
9755611938
add unittest for nearest_neighbor_interp_op
6 years ago
a24691a2a9
add nearest neighbor interpolation operator cpu kernel
6 years ago
8d3c3e048b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_reorg_op
6 years ago
d10b8efcc0
Merge pull request #14152 from Yancey1989/add_fused_broadcast_unittest
...
add fused broadcast op unit test
6 years ago
c21597cf07
fix(PE): use shared_ptr<BlockingQueue> for cross thread communication ( #14136 )
...
It seems that the blocking queue might be destroyed early than Run
method complete. It might because the Run method throw some unhandled
exception. However, it should be shared_ptr when multthread access an
resource. So change BlockingQueue as a shared_ptr.
test=develop
6 years ago
f3badacd97
Merge remote-tracking branch 'ups/develop' into fea/jit/gen
6 years ago
a53b1b0b1b
refine and init jitkernel vmul
6 years ago
2139b9f677
add jit gencode
6 years ago
06e508ab58
fix simple_on_word2vec random fail ( #14171 )
6 years ago
8899d42265
MKLDNN conv residual data: primitive reuse interface used. Reorder done when formats are different
...
test=develop
6 years ago
b73708d20b
add int and int64 dtype for gather_op ( #14175 )
...
test=develop
6 years ago
f11934cbe6
MKLDNN conv residual data: residual data is reorder when formats are incorrect
6 years ago
62a0fe0860
fix tensor array bug ( #14166 )
...
remove the optimized but buggy implementation
6 years ago
ed087f8232
refine op_handle ( #14178 )
...
test=develop
6 years ago
cdf2579d08
Merge pull request #14053 from jczaja/prv-seqpool-max
...
Max Sequence pool optimization
6 years ago
a3b26e8528
Merge branch 'develop' into grid_sampler
6 years ago
7333fe8e55
add math formula for exclusive/inclusive mode in avg pool. test=develop
6 years ago
35915fc543
Merge pull request #14147 from luotao1/remove_with_inference
...
remove with_inference option
6 years ago
90d9e5aee8
feat(platform): lazy initialization of devicecontext in pool ( #14067 )
...
* feat(platform): lazy initialization of devicecontext in pool
Use std::async(deferer, []{...}) to lazy initialize DeviceContext in Pool
test=develop
* Add future includes
test=develop
6 years ago
316765839d
add back jit simd instructions. stage.
6 years ago
eb7ed1b720
Merge pull request #13897 from gmcather/develop
...
1.add position encoding 2.logloss in nn.py
6 years ago
4e2aaf01bc
add depthwise conv mkldnn pass
...
added depthwise conv mkldnn pass which for MKLDNN changes depthwise_conv operator to conv operator because for mkldnn this is the same api
test=develop
6 years ago
fc23cc9d30
update paddle/fluid/API.spec
...
test=develop
6 years ago
6bfa6a0a33
add fused broadcast op unit test, test=develop
6 years ago
e2db0b9bf3
add a small test to verify tensor type
...
test=develop
6 years ago
bf2e4cb188
cleard. staged
6 years ago
70ce6dcd67
fix api_impl ci error ( #14140 )
6 years ago
eb37ed4c16
Merge pull request #14141 from JiabinYang/fix_inference_model_latest
...
Fix inference model not found on Mac CI
6 years ago
a943134a97
fix a few more tests
...
test=develop
6 years ago
2f639113ee
Fix sum_op's GetExpectedKernelType ( #14112 )
...
* fix sum_op's GetExpectedKernelType
test=develop
* fix ci fail
test=develop
6 years ago
5839e3236b
add program check
...
test=develop
6 years ago
ba22624d7e
position encoding && log loss
...
test=develop
6 years ago
3a96d41d72
remove with_inference option
...
test=develop
6 years ago
2494ca83ab
test=develop
6 years ago
ebfe5a02b3
merge develop branch
6 years ago
7c45e77c41
test=develop
6 years ago
b5f78ce42d
update paddle/fluid/API.spec
...
test=develop
6 years ago
cb27a9219d
Merge pull request #13971 from sefira/FasterOpDoc
...
generate proposal labels doc
6 years ago
5e5d2223a1
test=develop
6 years ago
3c957af139
Merge pull request #14080 from tensor-tang/refine/jit/crf2
...
Refine/jit/crf decoding
6 years ago
aa87a989ec
Merge pull request #14119 from Superjomn/fix/api-impl-tester
...
disable some tests
6 years ago
5f3acac9b3
update paddle/fluid/API.spec
...
test=develop
6 years ago
9ef19d4919
Merge pull request #14106 from luotao1/fix_cmake_warning
...
[1.1] fix cmake warning when ON_INFER=false
6 years ago
f2eed667c0
test=develop
6 years ago
16dfedb8b8
Merge pull request #14103 from jacquesqiao/cpu-for-1.1-merge-with-shape
...
[1.1] Cpu for 1.1 merge with shape
6 years ago
cef8cc81db
merge develop
6 years ago
458b16f42a
Rebase of seqpool-max optimization
...
test=develop
- Added rough profiling
- Profiled maxpool itself
- First draft of max seqpool optimization (is_test added)
- Added unit tests to seqpool
- Cosmetic fixes
- Fix to UT of Seq pool
Disabled grad checking for sequence max pool when is_test is set to True
-Cosmetic fix to comment
test=develop
- Fix to GPU build
test=develop
- yet another GPU fix for sequence max pool
- Fix to comment
test=develop
- Change to API of sequence_pool
test=develop
- Yet another API spec change
test=develop
6 years ago
5f7fda0b07
disable some tests
...
test=develop
6 years ago
ff6329bd5f
fix some inappropriate expressions in api doc for grid_sampler. test=develop
6 years ago
d3534d2b14
refine warning message
...
test=develop
6 years ago
177720a737
Merge pull request #14116 from chengduoZH/release/1.1.0
...
[1.1]Fix op_role value
6 years ago
acec4cb8ca
[1.1]fix op_role value
...
test=release/1.1
6 years ago
73671379cd
update paddle/fluid/API.spec
...
test=develop
6 years ago
8f1e398824
move param exclusive to the last in pool2d/pool3d for forward compatibility:. test=develop
6 years ago
593e1b18d7
fix some bugs and add some doc for GridSampleOp
6 years ago
0bb0e0c10f
add Grid Sampler Operator for STN.
6 years ago
3d4e050802
fix compile, optimize code test=develop
6 years ago
d26ff8cb2d
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpu-for-1.1-merge-with-shape
6 years ago
e0a89503f8
test=develop
6 years ago
0e3038680b
test=develop
6 years ago
79da263b11
Merge pull request #14032 from sfraczek/sfraczek/fix-test-multithreading-mkldnn
...
fix test resnet50 multi-threading on mkldnn
6 years ago
26200f2e42
[1.1] [project] train imagenet using large batch size ( #13766 )
...
* fix nccl2 lars dist support
* put lars in momentum op
* add tests lars
* fix ci
* fix cpu kernel
* soft warning
* remove lars in test_recognize_digits.py
* move to another op
* add file
* update api.spec test=develop
* update test=develop
* fix api.spec test=develop
* wip
* wip, finish grad merge ops
* wip, finish graph build
* wip test running
* work on 1 gpu
* workable version
* update
* fix tests
* fuse broadcast op
* fix compile failed
* refine
* add batch merge test mnist
* fix CI test=develop
* fix build
* use independent bn params for batch merge test=develop
* update api.spec
* follow comments and for test
* wip
* refine tests test=develop
* follow comments test=develop
* remove startup bn modify test=develop
* follow comments test=develop
* fix merge test=develop
6 years ago
2414f92f54
test=develop
6 years ago
8c1e304307
merge nn.py
6 years ago
45559d042c
move to pass
...
test=develop
6 years ago
c93e044ae0
add inclusive/exclusive mode in PoolOp avg pool type
6 years ago
9a74c4489f
test=develop
6 years ago
9dc28179a4
add similarity_focus op
6 years ago
7cd2417fe2
Merge branch 'develop' into cpu-for-1.1-merge-with-shape
...
test=develop
6 years ago
0a80f06ec4
Merge pull request #14086 from panyx0718/fix6
...
delete unused codes.
6 years ago
a314a80cdb
merge develop
6 years ago
4928ff32a9
fix cmake warning when ON_INFER=false
...
test=develop
6 years ago
c8adc2c6fe
cudnn version. staged.
6 years ago
06ffbc4f28
Merge branch 'shape_int_to_int64' of https://github.com/seiriosPlus/Paddle into cpu-for-1.1-merge-with-shape
...
test=develop
6 years ago
06de824ba8
fix shape in floats
6 years ago
ee74be3a49
[1.1] Bugfix/tensorarray ( #14044 )
6 years ago
33b4920d2d
Merge pull request #14057 from velconia/continue_hash_op
...
[1.1] Add hash_op implementation
6 years ago
209f24a241
Merge pull request #14051 from velconia/accelerate_embedding_grad
...
[1.1] Accelerate sparse embedding grad op in CPU device
6 years ago
2fec8c5d9a
Polish code
...
test=develop
6 years ago
7f7af5d412
Add xxhash deps to inference demo and trainer demo
...
test=develop
6 years ago
7cfc3c4415
Merge branch 'optimize-sum-seq-pooling-op' of ssh://github.com/jacquesqiao/Paddle into cpu-for-1.1-merge
6 years ago
72aef6b168
sum selected rows check empty
6 years ago
fe18adfbaa
Add fluid inference support
...
test=develop
6 years ago
c34610f86d
Fix lookup table at CPU Reduce strategy, test=develop
6 years ago
641369f92b
Merge branch 'dist-table-do-not-init-on-trainer' of ssh://github.com/jacquesqiao/Paddle into cpu-for-1.1-merge
6 years ago
d69c820707
Merge branch 'add-flag-to-control-rpc-thread-num' of ssh://github.com/jacquesqiao/Paddle into cpu-for-1.1-merge
6 years ago
1ed9ef6d70
Merge branch 'shape_int_to_int64' of https://github.com/seiriosPlus/Paddle into cpu-for-1.1-merge
6 years ago
f1a3fb041b
Merge branch 'fix_lookuptable_in_reduce' of https://github.com/seiriosPlus/Paddle into cpu-for-1.1-merge
6 years ago
da61a5b672
Merge branch 'optimizer-prefetch' of https://github.com/seiriosPlus/Paddle into cpu-for-1.1-merge
6 years ago
5ce3a32e06
Merge branch 'develop' into optimizer-prefetch
6 years ago
b6590b05fb
submit by tangwei12, test=develop
6 years ago
9da9b1926b
[1.1] fix graph num hang ( #14072 )
...
* fix graph num hang test=develop
* re-enable tests test=develop
* re-enable graph num check test=develop
* fix multi device pass role check test=develop
6 years ago
cb1ccc710b
fix shape type in uniform_random_op.cu
6 years ago
575f22711d
optimize code
...
test=develop
6 years ago
96d5500934
optimize code
6 years ago
748ee35c89
sum op handle empty input update selected_rows_functor.cu
6 years ago
dd78b5df93
sum op handle empty input
6 years ago
cbe128bbae
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into optimize-sum-seq-pooling-op
6 years ago
f4df0cb1a2
update the type of shape to int64, format code
6 years ago
fad42fe7cc
broadcast handle not inited parameter
6 years ago
7dcb0dc8c6
update year
6 years ago
68aeb4e7e9
add fake init test in test_dist_transpiler
6 years ago
5ed3e6f3f6
Merge pull request #14042 from luotao1/remove_unused_code
...
[1.1] remove unused code in paddle_inference_api.h
6 years ago
a13c788a04
fix a bug
6 years ago
97d47a7d08
Merge pull request #13913 from sneaxiy/seq_reverse
...
Add sequence_reverse_op
6 years ago
6e3615422f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_reorg_op
6 years ago
a3efba176c
Merge pull request #14085 from jerrywgz/fix_generate_proposals_op
...
[1.1] fix erase end in generate proposals op
6 years ago
7141debe38
add cudnn back. staged.
6 years ago
b9ae1c49f8
Merge pull request #13994 from guoshengCS/add-reshape-reuse-input
...
[1.1] Make reshape_op reuse input.
6 years ago
60058180cb
Merge pull request #13945 from sneaxiy/unify_mixed_vector_api
...
Unify API of mixed_vector in GPU and CPU
6 years ago
0328ffd3ab
add fake init op
6 years ago
bcc9126e7b
Merge pull request #14056 from panyx0718/fix
...
Fix threadpool
6 years ago
2098b42584
review fixes (Teamcity fails)
...
test=develop
6 years ago
961baea16c
Merge pull request #14063 from wojtuss/wojtuss/remove-unused-EnableMKLDNN
...
remove unused method from naive executor
6 years ago
379d933ae5
Merge pull request #14036 from phlrain/add_dropout_att_new
...
Add dropout att new 1.1 merge
6 years ago
d8b697357f
update height_sections to int64_t
6 years ago
a2820b9899
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_embedding_grad
6 years ago
bba0c4a9f2
delete unused codes.
...
test=develop
6 years ago
de2f965c9b
test=develop
6 years ago
cc0e23973d
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-reshape-reuse-input
...
test=develop
6 years ago
318ba99124
revert changes in protobuf.cc and type_defs
6 years ago
aa6dc82f4b
revert changes in protobuf.cc and type_defs
6 years ago
09409bad4d
staged. test speed=49ms in 1080.
6 years ago
64d5b4385e
fix crf decode avx512
6 years ago
21487d78bf
add crf decode jit kernel
6 years ago
b1fd62f39e
test=develop
6 years ago
3cfaeac288
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-reshape-reuse-input
...
test=develop
6 years ago
1af3fe8c35
test=develop
6 years ago
d5d09672c8
better fix
...
test=develop
6 years ago
de539d72da
format
...
test=develop
6 years ago
5be6f762d0
remove_lock_in_some_ops
...
test=develop
6 years ago
6c1d74bb47
Merge branch 'develop' into FasterOpDoc
...
test=develop
6 years ago
726fd438cd
avoid blocking everyone
...
please fix offline
6 years ago
7bcba47e41
test=develop
6 years ago
a7f94ec794
add similarity_focus op
6 years ago
8ab953e37c
auto insert infer_graph_clean_pass as the default first one
...
test=develop
6 years ago
d70c7fb9b3
Merge branch 'develop' into remove_unused_code
6 years ago
ea2bdd192d
Merge branch 'develop' into remove_unused_code
6 years ago
0de6811ee0
Change reserve to resize
...
test=develop
6 years ago
b58957d9d7
Revert "fix lookuptable in reduce strategy"
...
This reverts commit 0e722c5
6 years ago
9cad409f2a
test=develop
6 years ago
2761eafb92
shape type to int64_t, test=develop
6 years ago
d4a8967c1e
add const in &, test=develop
6 years ago
5660d6a3ba
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_embedding_grad
6 years ago
1f92c30565
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-reshape-reuse-input
...
test=develop
6 years ago
a05fce6544
Merge remote-tracking branch 'ups/develop' into fix/jit/avx
...
test=develop
6 years ago
bd064c0f44
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_reorg_op
6 years ago
0e25e397bd
shape type to int64_t, test=develop
6 years ago
d0fdcb2f6d
Merge pull request #14048 from velconia/change_sequence_pool_to_cpu
...
Accelerate Sequence Pool Grad Op
6 years ago
d1e85e33d7
shape type to int64_t, test=develop
6 years ago
d24d282a7a
fix avx error
...
test=develop
6 years ago
9cb8738f54
Merge pull request #14018 from tensor-tang/refine/jit/gru
...
Refine/jit/gru
6 years ago
70effddfc1
fix
...
test=develop
6 years ago
64e7688ade
clean more APIs
...
test=develop
6 years ago
c891bc22f5
clarify Reset
...
test=develop
6 years ago
6253b152e6
Merge branch 'optimize-sum-seq-pooling-op' of https://github.com/jacquesqiao/Paddle into optimize-sum-seq-pooling-op
6 years ago
14f5a40898
fix unit test
6 years ago
447a680a2b
Add API.spec
...
test=develop
6 years ago
5de4619781
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_embedding_grad
6 years ago
0695c1fbe8
Add remind for code
...
test=develop
6 years ago
0c5c4c4a5b
Add blas header file
...
test=develop
6 years ago
aac426444f
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-reshape-reuse-input
...
test=develop
6 years ago
d0ccdf8fc1
follow comments
...
test=develop
6 years ago
e2a348cd10
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into change_sequence_pool_to_cpu
6 years ago
f4e6fe0786
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into optimize-sum-seq-pooling-op
6 years ago
4f59690b4c
clean unused codes
...
test=develop
6 years ago
784a19ecd0
fix some thread-safty issue and simplify threadpool
...
test=develop
6 years ago
be58997443
remove unused method from naive executor
...
test=develop
6 years ago
40141f749b
Implement the unittest for hash op
...
test=develop
6 years ago
741cb33bd9
test multithreading
6 years ago
a53e8a8da6
Update MKLDNN integration framework to support Paddle multi-instances
...
Make all blob info saved in global device context to be thread based.
Meanwhile save thread id in thread local storage in ParallelDo
6 years ago
8a0f26f45f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into continue_hash_op
6 years ago
d4f9aa0852
Add hash op implementation
6 years ago
468467f391
update real incnet tester
6 years ago
39b3bf24d0
shape type to int64_t, test=develop
6 years ago
755927d2b0
shape type to int64_t, test=develop
6 years ago
7357d8412e
add flags for control the thead num for pserver
6 years ago
a4ad286e6b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_dropout_att_new
6 years ago
469bdb9e55
modify api.spec; test=develop
6 years ago
1a3b38a432
Polish code
...
test=develop
6 years ago
b154e0b492
clean demo_ci
6 years ago
133bac2b10
Accelerate embedding op grad
...
test=develop
6 years ago
2256fae45d
Merge pull request #13938 from NHZlX/ocr_attention_support
...
ceil pool mode support for ocr attention model.
6 years ago
abe8e207c4
clean demo_ci
6 years ago
597d92179b
clean demo_ci
6 years ago
201d4f2a85
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_dropout_att_new
6 years ago
a6e6bc45d6
modify dropout att; test=develop
6 years ago
2468057da6
Move code to SumSeqPoolGradFunctor
...
test=develop
6 years ago
9725db0d40
Fix copy wrong pos bug
...
test=develop
6 years ago
9c68709036
Accelerate sequence_pool functor
6 years ago
14ebc424d6
Add gpu support for unittest
6 years ago
e906c8e5e7
Merge pull request #14022 from jerrywgz/fix_rpn_target_assign_op
...
fix random fail in rpn target assign
6 years ago
bd5a82e193
Polish unit test code
6 years ago
047fa2f9aa
Add unit-test for sequence_pooling functor
6 years ago
c7379a7320
Fix top_k op ( #14034 )
...
1. Fix CUDA kernel when height is large than 2048.
2. Support input with more than 2D.
3. Fix unit test when k is large than 1.
4. Enhence unit testing.
test=develop
6 years ago
016bf51e3f
test=develop
6 years ago
f7bbcfa913
remove unused code in paddle_inference_api.h
...
test=develop
6 years ago
c056328563
test=develop
6 years ago
11f189bacf
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_demo_ci_trt
...
test=develop
6 years ago
8b7f45a889
add longs in framework
6 years ago
c13f1ef3c4
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_reorg_op
6 years ago
f3729db6e0
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into Pdv
6 years ago
8837669782
Merge pull request #13982 from panyx0718/fix
...
Clean up Reuse
6 years ago
dbd0075b68
Merge branch 'windows/support' into lb
6 years ago
c6dcffc61a
lb. add debug output
6 years ago
78cf76a1ca
fix linux compile
6 years ago
770e2a1881
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into Pdv
6 years ago
e943f4508b
add graph number check ( #14025 )
...
test=develop
6 years ago
92a2817a2b
test=develop
6 years ago
8e8e8e66ab
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_reorg_op
6 years ago
ae8f26072d
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_demo_ci_trt
...
test=develop
6 years ago
049c9c7d2a
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_dropout_att_new
6 years ago
ffb24a73ec
add dropout attr; test=develop
6 years ago
8f07f60915
test=develop
6 years ago
5993155d67
Merge remote-tracking branch 'dzhwinter/windows/support' into windows/support
6 years ago
f9e7cfb03c
save binary file
6 years ago
032c3a07e3
Merge remote-tracking branch 'ups/develop' into refine/jit/gru
...
test=develop
6 years ago
159be8cc63
optimize fusion gru kernel at size 8
6 years ago
607080e888
windows static library
6 years ago
23da8defc8
Merge pull request #14028 from luotao1/fix_resnet50_test
...
fix typo and warning in analyzer_resnet50_test
6 years ago
ff07dc315e
test=develop
6 years ago
a7497653d0
Refine Split op ( #13967 )
...
* speedup split_op
test=develop
* speedup split_op
test=develop
* rename ConcatGrad to Split
* refine concat and split
test=develop
* fix compile error
6 years ago
3099a8f3aa
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-reshape-reuse-input
...
test=develop
6 years ago
e0708e62ba
refine code
6 years ago
1c591c3909
Merge branch 'develop' into fix_rpn_target_assign_op
6 years ago
a9d7a9d720
test=develop
6 years ago
316bc9bfc9
fix typo and warning in analyzer_resnet50_test
...
test=develop
6 years ago
6447b69aec
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-reshape-reuse-input
...
test=develop
6 years ago
f06c6193d7
fix rpn target assign test=develop
6 years ago
563e7bca7f
"fix op. test=develop"
6 years ago
4625f83f92
better handle var type inference
...
avoid the default one that usually overwrites manually set ones
test=develop
6 years ago
8f2116d8fa
clean up after the changes have been stopped for so long.
...
test=develop
6 years ago
83dc689877
Merge remote-tracking branch 'ups/develop' into refine/jit/gru
...
test=develop
6 years ago
640e789d3d
add fusion gru jit kernel
6 years ago
39d39775c3
test=develop
6 years ago
70351de1b5
test=develop
6 years ago
0e24138494
Merge pull request #13991 from qingqing01/refine_generate_proposals_op
...
Refine generate proposals op
6 years ago
58c027cc38
Add rpc profiler flags. ( #13989 )
...
Add rpc profiler flags
6 years ago
d10e54c460
Merge pull request #14003 from chengduoZH/fix_fast_parallel_exe_bug
...
Fix test_parallel_executor_mnist.py randomly hang.
6 years ago
42aa1d409d
Merge pull request #13485 from tpatejko/tpatejko/capi-resnet-conv-elementwise-fusion
...
MKLDNN conv+elementwise_add fusion for residual connections in Resnet
6 years ago
664159ad42
Merge pull request #13998 from tensor-tang/fea/fusion_seqconv_add
...
Fea/fusion seqconv eltadd relu
6 years ago
6d3b030bb5
Refine the api of reshape to be compatible.
...
test=develop
6 years ago
82d2903b63
Fix fast ParallelExe bug
...
test=develop
6 years ago
aa35aaa1ab
MKLDNN conv + elementwise_add fusion: fixing formatting
...
test=develop
6 years ago
765085d297
Merge pull request #13904 from jerrywgz/roialign
...
Add RoI align operator.
6 years ago
56936b9e25
Refine doc for generate_proposals_op.
...
test=develop
6 years ago
ce2464fd98
MKLDNN conv + elementwise_add fusion: UT for missing bias added. UTs refactored. Some minor changes in the pass
6 years ago
4e72ab411e
MKLDNN conv + elementwise_add fusion: fix for crash when bias is not present
6 years ago
415b261555
MKLDNN conv + elementwise_add fusion: fusion options added
6 years ago
1676094697
MKLDNN conv + elementwise_add fusion: turn on residual connection pass when CAPI is used.
...
test=develop
6 years ago
0fe3079c46
MKLDNN conv + elementwise_add fusion: fix for order of parameters in elementwise_add in resnet50
...
test=develop
6 years ago
b73b868366
MKLDNN conv + elementwise_add fusion: bias in tests made persistent.
...
test=develop
6 years ago
a1fa203287
MKLDNN conv + elementwise_add fusion: name of the pass reused with name_scope_
6 years ago
2c43419db1
MKLDNN conv + elementwise_add fusion: comment explaining CorrectGraphEdges added
6 years ago
8fb29b2ca9
MKLDNN conv + elementwise_add fusion: new nodes marked as input or output
...
test=develop
6 years ago
cc1c8e37c1
MKLDNN conv + elementwise_add fusion: attributes in new conv op copied from old op
6 years ago
a27a8c5da8
MKLDNN conv + elementwise_add fusion: bias in test marked as persistable
6 years ago
af8c71317c
MKLDNN conv + elementwise_add fusion: CorrectGraphEdges refactored
6 years ago
3e033087f1
MKLDNN conv + elementwise_add fusion: LinkNodes function removed and
...
macro used.
test=develop
6 years ago
4be45af1cc
MKLDNN conv + elementwise_add fusion: skip connection attribute renamed. Comments about patterns added.
...
test=develop
6 years ago
9a335e0277
MKLDNN conv + elementwise_add fusion: changed a name of a formal argument in ElementwiseAdd pattern
6 years ago
fb7a50b230
MKLDNN conv + elementwise_add fusion: removed commented code. Internal functions marked as static.
...
test=develop
6 years ago
f688197182
MKLDNN conv + elementwise_add fusion: Fix output_data to point to the right tensor, also fix transpiler integration
6 years ago
efd76614fb
MKLDNN conv + elementwise_add fusion: implementation changed to conform with Paddle API
6 years ago
347bf90412
MKLDNN conv + elementwise_add fusion: bias is also handled
6 years ago
bf95ac36a7
MKLDNN conv + elementwise_add fusion: further reformatting
6 years ago
cbe122ae2e
MKLDNN conv + elementwise_add fusion: correcting formatting
6 years ago
2a251bbf27
MKLDNN conv + elementwise_add fusion: some refactoring: consts, function calls instead of constant values
6 years ago
b8e54ab5cc
MKLDNN conv + elementwise_add fusion: parameter name changed to ResidualData
6 years ago
27573ece03
MKLDNN conv + elementwise_add fusion: trailing spaces removed
6 years ago
7f5c8a95e8
MKLDNN conv + elementwise_add fusion: arguments are replaced for many parameters in operator
6 years ago
5996bd39e8
MKLDNN conv + elementwise_add fusion: graph is corrected based on actual argument name, not formal argument name
6 years ago
41f3d78fdf
MKLDNN conv + elementwise_add fusion: output and elemwise param share data in conv primitive. Output is properly allocated
6 years ago
07a62ddc08
MKLDNN conv + elementwise_add fusion: inputs in pass modified. Support for new conv parameter. UTs corrected
6 years ago
56528531ea
MKLDNN conv + elementwis_add fusion: initial work on passing eltwise data to conv primitive
6 years ago
42f569fdfd
MKLDNN conv + elementwise_add fusion: use_mkldnn attribute added
6 years ago
441d3a4726
MKLDNN conv + elementwise_add: added some refactoring in the pass
6 years ago
38b7b34b1c
MKLDNN conv + elementwise_add fusion: added reachability tests, inputs and outputs in graph nodes are transformed
6 years ago
16eaaf3fbe
MKLDNN conv + elementwise_add fusion: added one more UT, found and corrected bugs in pass
6 years ago
604bad08bc
MKLDNN conv + elementwise_add fusion: implementation of patterns refarctored, applied to graph. UTs added
6 years ago
9ce343f868
MKLDNN conv + elementwise_add fusion: initial implementation of patterns
6 years ago
40f8456a4f
refine fuse pattern and attr
...
test=develop
6 years ago
cbbacb2534
Merge remote-tracking branch 'ups/develop' into fea/fusion_seqconv_add
...
test=develop
6 years ago
603ba5e01d
add seqconv eltadd relu pass
6 years ago
4801ee8f97
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_generate_proposals_op
6 years ago
da722d6d9b
Merge pull request #13858 from Sand3r-/mgallus/conv-bias-pass
...
[MKLDNN] Fuse Conv + Bias using Pass
6 years ago
a4b48f70c1
Merge pull request #13997 from wojtuss/wojtuss/do-not-enable-mkldnn-twice
...
do not enable MKL-DNN twice
6 years ago
252401c5d3
Merge pull request #13992 from wojtuss/wojtuss/add-ifdef-mkldnn
...
Add ifdef guard for MKL-DNN placement pass
6 years ago
f9ca31811d
Remove use mkldnn from config in resnet50 test
...
test=develop
6 years ago
23fc896bc2
Merge remote-tracking branch 'ups/develop' into fea/fusion_seqconv_add
...
test=develop
6 years ago
339e655aec
refine and add seqconv elementwiseadd relu op test
6 years ago
c504a5a1b7
Adjust Conv+bias to placement pass
...
test=develop
6 years ago
d7509d63f1
Conv+Bias: Support non-null bias
...
test=develop
6 years ago
91e8fbac2f
Enable MKLDNN in Resnet50Tester
...
test=develop
6 years ago
582f59c190
Conv+Bias fuse
6 years ago
a1d3db031b
Merge pull request #13844 from jerrywgz/fix_roi_pool
...
fix roi pool register
6 years ago
dfb841ad5a
Make reshape_op reuse input.
...
test=develop
6 years ago
8e0b9496de
Fix unit test
...
test=develop
6 years ago
e6f480ec44
add comment on the default first pass
6 years ago
2cf258e381
remove redundant pass list
6 years ago
5632019f0f
add MKL-DNN placement pass
...
This patch also refactors conv+bn (includes changes from PR
https://github.com/PaddlePaddle/Paddle/pull/13926 )
updated to use the mkldnn-placement-pass.
test=develop
6 years ago
0a9f5f1790
Merge pull request #13968 from tensor-tang/fix/jit/exp
...
Fix jit exp
6 years ago
5083ec3a1b
do not enable MKL-DNN twice
...
After the MKL-DNN placement pass there is no need to enable MKL-DNN
in operators via executor
test=develop
6 years ago
fcb2e8103e
Ocr end2end dev ( #13889 )
...
* add detect and end2end code
* update the scale for coodinates restore
* fix merge bug with dev.
* fix merge bug with dev.
* test=develop
* fix code style test=develop
* fix code style test=develop
* test=develop
* test=develop
* test=develop
6 years ago
e5ce965952
refine and add eltadd_relu unit test
6 years ago
5a38930660
test=develop
6 years ago
c3b70aece9
Add MKL-DNN placement pass ( #13958 )
...
* add MKL-DNN placement pass
This patch also refactors conv+bn (includes changes from PR
https://github.com/PaddlePaddle/Paddle/pull/13926 )
updated to use the mkldnn-placement-pass.
test=develop
* remove redundant pass list
* add comment on the default first pass
* fix test for conv+relu mkldnn fuse
6 years ago
4a368a4901
add ifdef guard for MKL-DNN placement pass
...
test=develop
6 years ago
909e1341bd
Merge pull request #13966 from panyx0718/fix4
...
set default timeout to avoiding blocking CI
6 years ago
9775e50ca2
Fix add doc for bias_attr ( #13937 )
...
* fix conv doc
test=develop
* fix seq_conv doc
test=develop
* fix simple_img_conv_pool
test=develop
* update API.spec
* update parameter doc
test=develop
* follow comment
test=develop
* fix other layer
test=develop
* fix lstm bias_attr doc
test=develop
6 years ago
7b11162ab5
Merge pull request #13949 from PaddlePaddle/wojtuss/unique-patterns-request-comment
...
Added comment with request for enhancement
6 years ago
7cb19a5976
fuse elementwise_add and relu
6 years ago
3c249283af
init seqconv eltadd relu op
6 years ago
48982e9dc7
fix lookuptable in reduce strategy
6 years ago
9a819265eb
fix
...
test=develop
6 years ago
ac2eba4457
test=develop
6 years ago
305034f5b3
Merge pull request #13909 from luotao1/mkldnn_test
...
refine mkldnn test in analyzer_tests
6 years ago
b77e4f4978
update
...
test=develop
6 years ago
553342624e
test=develop
6 years ago
9a14ca91b8
test=develop
6 years ago
60ff05e312
Merge branch 'luotao1-fix_rnn2_test' into fix/jit/exp
...
test=develop
6 years ago
ef09862450
fix analyzer_rnn2_test
...
test=develop
6 years ago
0e722c5ea2
fix lookuptable in reduce strategy
6 years ago
e5b4643ad8
add profile_mkldnn test
...
test=develop
6 years ago
7d680be5a3
Merge branch 'develop' into mkldnn_test
6 years ago
0bb3b099c2
generate_proposal_labels doc
6 years ago
55fd136ab0
Added comment with request for enhancement
...
This adds a `TODO` comment according to https://github.com/PaddlePaddle/Paddle/issues/13550#issuecomment-430133585
test=develop
6 years ago
a831ecc75d
Add grpc error context. ( #13957 )
...
Add grpc error context
6 years ago
b139b687de
Merge remote-tracking branch 'ups/develop' into fix/jit/exp
...
test=develop
6 years ago
Wu Yi
Kaipeng Deng
whs
sneaxiy
tangwei12
dengkaipeng
Tao Luo
Yan Chunwei
chengduo
dzhwinter
tensor-tang
Xin Pan
JiabinYang
Yan Xu
Yu Yang
Tomasz Patejko
Sylwester Fraczek
barrierye
Yancey1989
gmcather
qingqing01
Jacek Czaja
dengkaipeng
Qiao Longfei
Qiyang Min
minqiyang
Zeng Jinle
Guo Sheng
Hongyu Liu
jerrywgz
guosheng
buxingyuan
Wojciech Uss
Brian Liu
phlrain
Zhaolong Xing
gongweibao
Michal Gallus
Michał Gallus
Yipeng