5d4d117edc
Merge pull request #14502 from qingqing01/cudnn5_fix
...
Fix compling with cuDNN v5
7 years ago
f7b55de9e5
Merge branch 'develop' into enhance_hierachical_sigmod_op
7 years ago
e68c1fcd5a
Merge pull request #14522 from reyoung/feature/fix_op_header_deps
...
fix(Compile): fix depends error when compile op using cub
7 years ago
6eba5bd276
Fix direct copy and refine split ut
...
test=develop
7 years ago
fd290c2580
fix mac compile of analysis
...
test=develop
7 years ago
5857fb3014
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
...
test=develop
7 years ago
3562051302
add gru refer code and remove redundant avx code
...
test=develop
7 years ago
af9a3301da
test=develop
7 years ago
3e3599f3d9
Refine split tensorrt plugin
7 years ago
f10e196fc8
fix build issue
7 years ago
6a128dea32
Merge pull request #14515 from reyoung/feature/fix_macos_build
...
fix(Macos): fix compile on macos
7 years ago
ad349e770f
Merge pull request #14452 from NHZlX/fix_avg_pool_trt_bug
...
fix avg pool trt bug
7 years ago
f913860873
jitkernel lstm refer support peephole
...
test=develop
7 years ago
2f9b5f2383
Merge branch 'develop' into fea/jit/rnn
7 years ago
014e50c284
test=develop
7 years ago
6e66fadb95
clean up the pre-definitions on windows
7 years ago
3edd32d070
fix(Compile): fix depends error when compile op using cub
...
some operators depend on cub and xxhash by header. The dependency should be declared explicitly rather than declared to pybind.
test=develop
7 years ago
cda60311f9
Fix compling with cuDNN v5
...
test=develop
7 years ago
67562a6fcd
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
703b26e697
add profiler, parallel_executor back
7 years ago
1d9b2a453c
Merge pull request #14508 from luotao1/warm_up_multi_thread
...
add warm up in TestMultiThreadPrediction
7 years ago
b3364d4035
fix(Macos): fix compile on macos
...
test=develop
7 years ago
a685f305f8
Merge pull request #14479 from reyoung/feature/fix_macos_ut
...
fix(Mac): fix unittest of macos
7 years ago
10fb4ceefc
Merge pull request #14351 from tpatejko/tpatejko/mkldnn-elementwise_mul
...
[MKLDNN][JIT][AVX512] Elementwise Mul
7 years ago
13e254faed
refine code, test=develop
7 years ago
b4c826c548
Merge remote-tracking branch 'ups/develop' into fea/jit/rnn
...
test=develop
7 years ago
ce31deb7e9
refine refer code and add lstm refer code
...
test=develop
7 years ago
79cec53111
add ignore index for sigmoid cross entropy with logits op, test=develop
7 years ago
e62872df8b
fix conflicts
7 years ago
a4dc1d4292
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_trt
...
test=develop
7 years ago
faeb9b8aa9
fix compile rely problem
7 years ago
a8d3aaae2a
print output log warning ( #14497 )
...
test=develop
7 years ago
eb9b9becdc
add warm up in TestMultiThreadPrediction
...
test=develop
7 years ago
c2cfb03a72
add lstm jitcode
7 years ago
5cc7946313
Merge pull request #14499 from luotao1/disable_openblas_test
...
disable two openblas test temporary
7 years ago
10ae3ba486
Merge pull request #14493 from hjchen2/develop
...
Implement leaky relu converter from fluid to tensorRT
7 years ago
2a84054372
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_trt
...
test=develop
7 years ago
b742d46520
fix demo ci bug on trt
7 years ago
25adf970b2
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
33c65517fd
Update CMakeLists.txt test=develop
7 years ago
1d3e9bde1e
Merge pull request #14488 from yihuaxu/develop_7a64d48f5_stack_opt
...
Optimize the stack operator
7 years ago
01bda73116
Update CMakeLists.txt
7 years ago
09ee266f8e
disable two openblas test temporary
...
test=develop
7 years ago
2c2a192eb1
Resolve merge conflicts
...
test=develop
7 years ago
8bc1c5d2ab
Implement the Tensorrt plugin for elementwise op ( #14487 )
...
* Initialize the elementwise plugin.
* Implement the basic CUDA kernel of elementwise plugin.
test=develop
7 years ago
7aa3aff338
Merge pull request #14465 from tensor-tang/fea/jit/exp
...
jitcode act support all size
7 years ago
1b894e495f
Merge pull request #14437 from jczaja/prv-softmax-mkl
...
Introducing MKL to softmax for inference
7 years ago
3f73c0a70d
fix the build issue on windows
7 years ago
a94a7355f0
Refine the GraphNum check ( #14144 )
...
* refine GraphCheck
test=develop
* fix ci fail
test=develop
7 years ago
3a72a634cf
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
a906a361be
Add the macro for NVCC (test=develop)
7 years ago
d91740acb1
Revert "Remove the remnant code (test=develop)"
...
This reverts commit be50670348
7 years ago
be50670348
Remove the remnant code (test=develop)
7 years ago
1622cb9937
Fix alpha tensor key
7 years ago
a8c077df7c
Implement leaky relu tensorRT converter
7 years ago
9eefd2c766
Modify some infer-shape about detection operators in compile-time. ( #14483 )
...
* Modify some infer-shape in compile-time.
7 years ago
cf685f361b
Merge pull request #14458 from tpatejko/tpatejko/mkldnn-skip-connections
...
[WIP] Correcting and extending MKLDNN residual connection fuse pass
7 years ago
f4c869d872
Optimize the layer_norm operator with AVX intrinsic function ( #14417 )
...
* Optimize layer_norm operator with AVX intrinsic functions
* Revert the wrong modifications
* Implement the jit kernel for layer_norm operator
* Add math headfile to fix the compile issue (test=develop)
* Add math headfile to fix the compile issue (test=develop)
* Fixed the intrinsic headfile issue (test=develop)
* Fix the conflicts (test=develop)
* Revert for CUDA compiler (test=develop)
* Fixed the cuda depency (test=develop)
* Fix the marco issues (test=develop)
7 years ago
816b464037
Merge pull request #14486 from hjchen2/develop
...
Fix tensorrt plugin cmake dependency, test=develop
7 years ago
81f750a88c
fix the dependency
7 years ago
ee0fd78c81
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
f1a392a5fe
Merge pull request #13804 from sneaxiy/rewrite_allocation
...
Rewrite allocation
7 years ago
f418f552df
Merge branch 'develop' into develop_7a64d48f5_stack_opt (test=develop)
7 years ago
8443961a4f
add warp_ctc back
7 years ago
2825685f2a
Fix tensorrt plugin cmake dependency, test=develop
7 years ago
e878a8e885
update
...
test=develop
7 years ago
fd7e643153
Convolution fusion operator. ( #14449 )
...
* Convolution fusion operator.
* Clean code
test=develop
7 years ago
98bbfc17be
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rewrite_allocation
...
test=develop
7 years ago
7486b0ddec
fix(Mac): fix unittest of macos
...
test=develop
7 years ago
4a6769da84
re-organize the cmake file
7 years ago
8ef6280c03
Add operator double support. test=develop
7 years ago
d424115f9e
Clean code
...
test=develop
7 years ago
1aff40a4c6
exclude warpctc_op on windows
7 years ago
7d51a0e887
disable DSO by default on windows
7 years ago
b967e01cbe
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
d7bd0361cb
fix dist deps ( #14471 )
...
* fix dist deps test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
7 years ago
b12c77dae2
Fix unittests
...
test=develop
7 years ago
9b0eae3023
- Removing partial specialization of sotmax for inference for GPU
...
test=develop
7 years ago
c59d3e83bc
test case fix
7 years ago
a3e952f41d
add the jit back
...
fix compile error on windows
7 years ago
a19b3225a1
fix jitcode small size
...
test=develop
7 years ago
be80bb4f28
- Fix to GPU
...
test=develop
7 years ago
4dbdfa60ef
sigmoid and tanh support all size
...
test=develop
7 years ago
ccb8963705
refine exp jitcode with all size
...
test=develop
7 years ago
1cc23ef67d
merge from paddle:develop
7 years ago
d3eae8f61b
refine relu and fix addrelu test
7 years ago
4e67fe6a12
refine act and vxx with all size
7 years ago
ba3eaed7a7
exp support all size
7 years ago
1ffce8c0ae
fix build error on noavx
...
test=develop
7 years ago
4bf6817cbc
fix gpu load model
...
the parameters will load from CPUPlace, that will keep copying data
between CPU and GPU places.
test=develop
7 years ago
c69c41604e
MKLDNN elementwise_mul: Move Kernel to KernelPool to avoid segfaults
...
test=develop
7 years ago
785066eb8a
MKLDNN elementwise_mul: Check if AVX512 is available
...
test=develop
7 years ago
08f63c4d12
MKLDNN elementwise_mul: Lint changes to UT & integration
...
test=develop
7 years ago
49b09327f6
MKLDNN elementwise_mul: Reorder on non-nchw input, fallback on non-16 divisable fm
...
test=develop
7 years ago
d14858e4ba
MKLDNN elementwise_mul: Parallelize mul
7 years ago
ed31936ba1
MKLDNN elementwise_mul: Support NCHW, update UT
7 years ago
4e54ab76ec
Add HasAttr method to Operator
7 years ago
700bcbf74f
MKLDNN elementwise_mul: h and w loops implemented in xbyak
7 years ago
ad09facafe
MKLDNN elementwise_mul: CPU tests initially refactored. MKLDNN mul test for broadcast added
7 years ago
2d73ad180a
MKLDNN elementwise_mul: simple xbyak version for AVX512
7 years ago
213ec37d6a
MKLDNN elementwise_add: simple initial implementation of the operator for MKLDNN format
7 years ago
a2d9b34417
Refine operator cmake ( #14413 )
...
* wip simplify operator framework
* wip
* wip
* done test=develop
* clean test=develop
* fix test=develop
* fix deps test=develop
* fix cpu build test=develop
* fix tensorrt build test=develop
* fix tests test=develop
* fix test=develop
* fix cpu build test=develop
7 years ago
53da846d1e
MKLDNN residual connections fuse pass: initial implementation of fusion for projection pass
...
test=develop
7 years ago
764f97deac
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
8580b7a130
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
7f17e561d7
Merge pull request #14423 from tensor-tang/fea/jit/act
...
jitcode act relu, exp, sigmoid, tanh
7 years ago
28bd5b7bad
fix space_to_depth_op unicode problem ( #14430 )
...
* fix space_to_depth_op unicode problem
* test=develop
7 years ago
513bb6c151
Squashing MKL based softmax for inference
...
test=develop
- Added profiling to softmax functors
- MKL based softmax inference op
- Fix to softmax compuation via MKL
- cleaning
- Cosmetic fixes to softmax MKL
- Fix to ON_INFER lack of propagation
7 years ago
9b64aac41f
add macro for pool2dDirectCUDAFunctor
...
test=develop
7 years ago
dbc4fcd722
MKLDNN residual connections fuse pass: unit tests enabled and added
7 years ago
4224089354
MKLDNN residual connections fuse pass: Maybe removed and boost::optional used where it makes sense
7 years ago
86fd3b32be
MKLDNN residual connections fuse pass: counting statistics added to the pass
7 years ago
ee6f778beb
MKLDNN residual connections fuse pass: further refactoring
7 years ago
7423748e37
MKLDNN residual connections fuse pass:
...
* implements reachability check between identity node and non-identity argument to elementwise_add
* implements handling identity node as x and as y argument to elementwise_add
7 years ago
8f9a8c455a
delete unused test code.
...
test=develop
7 years ago
1722678258
Make nce support more distribution. ( #13549 )
...
* Fix truncated normal.
* Fix.
* Make nce support more distribution.
* Fix API.spec.
* Fix python API.
* Fix.
test=develop
* Fix API.spec
test=develop
* Fix sampler.
* Fix order of arguments in python API.
test=develop
7 years ago
83f8c403a7
Merge branch 'develop' of https://github.com/paddlepaddle/paddle into fix_avg_pool_trt_bug
...
test=develop
7 years ago
b969116988
fxi avg pool trt bug and fix cpplint
7 years ago
1f00723fa3
exp, sigmoid, tanh jitcode support more size
...
test=develop
7 years ago
19e669a992
Add legacy_allocator
...
test=develop
7 years ago
2f27c048cc
Merge pull request #14440 from hjchen2/develop
...
Add PRelu tensorRT plugin and Conv2d transpose op converter
7 years ago
1cb7e7dda2
fix(allocation): fix ut
...
test=develop
7 years ago
d971d5b875
Merge pull request #14431 from velconia/fix_expand_op_dim_in_compile_time
...
Fix expand op incorrect infer shape
7 years ago
6a7b995737
Refine commit message to enable ci, test=develop
7 years ago
b32c13dc20
Add cudnn ctc loss ( #12366 )
...
* add cudnn ctc loss
* wip add test test=develop
* wip
* wip
* done test=develop
* move include cudnn test=develop
* test test=develop
* fix build test=develop
* fix build test=develop
* fix build on cudnn5 test=develop
* fix cudnn5 build test=develop
* fix cudnn5 build test=develop
* merge develop softmax functor change test=develop
7 years ago
d1a1fafc4c
code style
7 years ago
8cda7b3d20
Merge remote-tracking branch 'ups/develop' into fea/jit/act
...
test=develop
7 years ago
e2d6eddd32
remove ComputeDeprecated
...
test=develop
7 years ago
6d0d5a76eb
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
162f2d4109
disable the openblas multi-thread on windows since no support
...
adjust the python script
7 years ago
7796f65f89
fix inference on gpu out of mem ( #14414 )
...
* fix inference on gpu out of mem
the transfer logic in operator.cc will keep creating new scopes.
7 years ago
f115eb0d1e
enhance api. test=develop
7 years ago
64f7516aee
fix lrn on mac ( #14426 )
...
* rename and fix blas vsqr
test=develop
* update
7 years ago
c8f6e70ab4
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rewrite_allocation
...
test=develop
7 years ago
413f5948b2
Fix code style
7 years ago
21f33b4274
Complete PRelu plugin and Conv2d transpose op converter
7 years ago
f65ddff8d1
unify act jitcode of relu, exp, sigmoid and tanh
7 years ago
6a159071b6
add vtanh jitcode of size 8
7 years ago
046374bcd1
add vsigmoid jitcode of size 8
7 years ago
560b29ccb7
Polish code
...
test=develop
7 years ago
21d6e8e8c8
Polish code
...
test=develop
7 years ago
50b6e4c6bc
Fix expand grad op infer shape
...
test=develop
7 years ago
8a1eeec579
add mkldnn prop_kind phase for inference-only case to pooling and activations ( #14278 )
...
* add is_test to pooling and activations
add prop_kind support for layers activation. conv and pooling
add a pass that sets is_test to true
add transpiler version of is_test pass
test=develop
* patch test and pass
test=develop
* add pass to analyzer.h
test=develop
* add is_test attr description & pass only on mkldnn
in:
activation_op.cc
batch_norm_op.cc
conv_op.cc
dropout_op.cc
lrn_op.cc
pool_op.cc
sequence_pool_op.cc
softmax_op.cc
* fix is_test handling for activation pool and conv
* change description of is_test for all layers again
* remove GetAttr(use_mkldnn) from pass
* rename correct_mkldnn_test_phase to is_test
and remove dependency on MKLDNN
test=develop
* review fix magic number
* two if(..)s into one
* Check is_test once and pass mkldnn forward prop kind
* dereference shared_ptr with * (without get())
test=develop
* add is_test_pass back
test=develop
7 years ago
d1429ac4a5
add recordio support
7 years ago
82773477ae
Add selu ( #14415 )
...
* add selu
* use for range
test=develop
* add API
test=develop
* follow comment
test=develop
* update API.spec
test=develop
7 years ago
95d5060ddd
fix abs -> fabs error. test=develop
7 years ago
9d29ebc010
Merge pull request #14306 from sfraczek/sfraczek/test-analyzer-mobilenet
...
add test_analyzer_mobilenet
7 years ago
30147d7f58
Fix expand op incorrect infer shape
...
test=develop
7 years ago
d318583eb5
rename mobilenet dir to mobilenet_depthwise_conv
...
test=develop
7 years ago
ba9ff508e8
temp fix
7 years ago
e5c4cf6140
Polish allocation
...
Clean allocation->Deleter
test=develop
7 years ago
03ccb9a461
Optimize the stack operator
7 years ago
2faa2b4048
remove cu file. test=develop
7 years ago
ee2a7f1b8c
refine exp and fix error on avx
...
test=develop
7 years ago
1e06a32a0d
add vexp jitcode of size 8
...
test=develop
7 years ago
2354409601
Merge pull request #14374 from tensor-tang/fea/jit/act
...
add vrelu jitcode
7 years ago
0d6718fcbd
Pass compile
7 years ago
1d867805b0
rollback analyzer_seq_conv1_tester
...
test=develop
7 years ago
5ef123c778
Merge branch 'develop' into dam_fc
7 years ago
d3aed98d86
Merge pull request #14320 from wopeizl/windows/online
...
Windows/online
7 years ago
9e6b1c5f97
Refine tester of TensorRT engine ( #14390 )
...
* Refine the tester for MixedRTPredictor.
test=develop
* Enable the profiler in TensorRT engine.
* Support the use of combined inference model in TensorRT unittest, and print the shape of feed targets.
7 years ago
d3e63e6e04
Merge pull request #14412 from jczaja/prv-dam-softmax
...
Softmax for Inference is enabled when ON_INFER is set
7 years ago
77ac30e5fa
Merge pull request #14386 from NHZlX/add_trt_plugin
...
add plugin support for paddle-trt
7 years ago
0ef2a37c0e
merge from develop
7 years ago
be332a13bc
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
8cfda7ee0c
Merge pull request #14382 from panyx0718/fix4
...
Refine the pass builder and buildstrategy
7 years ago
b361579f09
- Softmax for Inference is enabled when ON_INFER is set
...
test=develop
7 years ago
980a6753a8
fix typo to pass the ci
...
test=develop
7 years ago
8f301f4618
Merge pull request #14381 from qingqing01/manylinux_v5_fix
...
Fix compiling with cuDNN v5.
7 years ago
15bdb7ef14
delete error uploaded files
...
test=develop
7 years ago
d93b2d0365
Refine code
7 years ago
2412c27c2b
Merge branch 'develop' into sfraczek/test-analyzer-mobilenet
7 years ago
c7b3bfcdf1
Merge pull request #14376 from baojun-nervana/intel/ngraph_fusedop
...
Adding fused operator for ngraph
7 years ago
1a9008c420
code style fix
...
test=develop
7 years ago
e0d4e04bdd
fix some compiler warning
...
test=develop
7 years ago
ea81f8eed2
Clean interface of allocator
...
Clean managed/umnamaged allocator
7 years ago
8ea13e336a
add in_num_col_dims for fc
7 years ago
bae3659714
more test
...
test=develop
7 years ago
ddb120357c
Merge branch 'develop' of https://github.com/paddlepaddle/paddle into add_trt_plugin
...
merge develop and fix conflicts
7 years ago
a507845a77
test=develop
7 years ago
2abee2091b
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
9f33593910
human readable memory warns ( #14361 )
...
* human readable memory warns test=develop
* update test=develop
* refine test=develop
* fix build test=develop
7 years ago
9eb0ab1db3
Merge pull request #14384 from tensor-tang/refine/lrn
...
Refine lrn cpu forward
7 years ago
08d1dc84a9
fix
7 years ago
42c48c3a82
fix
7 years ago
447bf7c80b
test=develop
7 years ago
203ec852cf
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
30ddc07a7e
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
a61909ff47
test=develop
7 years ago
e65cbd3b06
Merge pull request #14387 from jacquesqiao/lookup_sparse_table_add_test_mode
...
Lookup sparse table add test mode
7 years ago
6cf8f24b1b
Merge pull request #14389 from jacquesqiao/fix_sgd_op_optimize_sparse_table
...
sgd_op optimize selected rows do not enforce id < height
7 years ago
7066b3850a
Merge pull request #14395 from sneaxiy/fix_num_threads_in_fast_pe
...
Fix num_threads settings in fast_pe
7 years ago
10ab177f89
Merge pull request #14403 from PaddlePaddle/revert-14337-prv-dam-softmax
...
Revert "Softmax op optimization for inference "
7 years ago
9f252e0032
Combine Inference Analysis with IR ( #13914 )
7 years ago
5b9c62faee
Revert "Softmax op optimization for inference "
7 years ago
51a538e055
Fix style and use enum
...
test=develop
7 years ago
6490bb2765
Merge pull request #14337 from jczaja/prv-dam-softmax
...
Softmax op optimization for inference
7 years ago
0b96268057
fix comments
...
test=develop
7 years ago
38d32c98b8
merge develop
...
test=develop
7 years ago
eb18d532a5
fix num_threads in fast_pe
...
test=develop
7 years ago
9f68e9a7fe
fix auc op ( #14385 )
...
test=develop
7 years ago
a0284f6fbc
Add backward CPU kernel. test=develop
7 years ago
d219818434
Fix compiling in cuDNN v5.
...
test=develop
7 years ago
efb5c03f60
sgd_op optimize selected rows do not enforce id < height
...
test=develop
7 years ago
51f3838f96
add log for not exist code
...
test=develop
7 years ago
7aa8b2ccf2
optimize code
7 years ago
db06568e69
test=develop
7 years ago
e5bf8616f0
Merge branch 'develop' of https://github.com/paddlepaddle/paddle into add_trt_plugin
...
test=develop
7 years ago
d38fd6a0fc
add plugin support and offer an simple split sample
7 years ago
8d205c853c
add is_test for lookup_sparse_table
7 years ago
b4dfba1779
refine lrn_op cpu forward and speedup
...
test=develop
7 years ago
1be85d011d
add mkl vsqr and vpow
7 years ago
f4be1d99d0
polish code and test
7 years ago
6c7b64cc20
Support softmax return in softmax_with_cross_entropy ( #14367 )
...
* Support softmax return in softmax_with_cross_entropy
* Add test for return_softmax=False
test=develop
7 years ago
ea3538d8dd
Added fused operator
...
test=develop
7 years ago
759ffca423
some improvements
...
test=develop
7 years ago
99dffb91d6
allow to repeatedly share and update BuildStrategy
...
test=develop
7 years ago
6c32945556
Merge pull request #14372 from luotao1/speedup_analysis
...
speedup DetectPatterns
7 years ago
4a55fb5f5b
Add density_prior_box_op ( #14226 )
...
Density prior box operator for image detection model.
7 years ago
2d7134bc37
add initial code for plugin
7 years ago
0043c42b3e
add vrelu jitcode
...
test=develop
7 years ago
36c46152e1
Add unittest for yolov3_loss. test=develop
7 years ago
77c1328fa7
add CPU kernel forward
7 years ago
5d0b568ecb
Add YOLOv3 loss operator. test=develop
7 years ago
668ae523d2
speedup DetectPatterns
...
test=develop
7 years ago
b8ff0972b6
test=develop
7 years ago
32e05b01f2
test=develop
7 years ago
02631965c8
Refine
7 years ago
9a6e239281
fix mac graph detector sort ( #14356 )
7 years ago
c27554ac33
Merge pull request #14336 from jacquesqiao/add_bilinear_tensor_product_layer
...
add bilinear_tensor_product layer
7 years ago
991a9f7b72
Merge pull request #14358 from NHZlX/add_serial_and_filter_logs
...
Set serial for tensorrt utest and set unused logs invisible.
7 years ago
bb3f6bd31c
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
1b75fd2236
revert
7 years ago
61fa5218b9
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
bd2943788b
Fix gather & stack op ( #14355 )
...
* Add int type support for stack_op
* Improve gather op to support index with shape N x 1
test=develop
* Fix stack_op kernel's registry
test=develop
7 years ago
8f9bfad246
perf(compile): speed up reduce_op compile by splitting files ( #14294 )
...
test=develop
7 years ago
397de907ed
merge develops
...
test=develop
7 years ago
d6ff006903
add serial to trt test and do not print log for unused trt logs
7 years ago
13bfee1f85
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
7840d181c9
fix style issue
7 years ago
dc339b78d7
fix code style
7 years ago
d231e55065
merge develop
...
test=develop
7 years ago
cf8d2e67e3
clean buffered_allocator
7 years ago
ef8a7db81e
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
c8801e100f
grad diff problem to be fixed and need api spec change to be done
7 years ago
03299ed46c
- Fix to linking for GPU builds of softmax inference
...
test=develop
7 years ago
0756343767
- Fix GPU compilation
...
test=develop
7 years ago
d332326847
- Added unit tests for softmax is_test=True op
...
test=develop
7 years ago
c1fccc29c1
- Noise adding removed for Test phase of softmax
7 years ago
573e68eb40
Merge pull request #14348 from luotao1/speedup_analysis
...
skip mkldnn related pass when use_mkldnn=false to speedup analysis
7 years ago
9b558a8035
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
7638f0afb3
simplify the logic
7 years ago
d01a26280e
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
ff28b1ffc0
Merge pull request #14071 from barrierye/add_similarity_focus_op
...
Add similarity focus op
7 years ago
688ed60116
Add lod tensor array to tensor op ( #13990 )
...
* add lod tensor array concat
* add lod tensor array concat
* test=develop
* add lod tensor array concat
test=develop
* Fix API.spec
test=develop
* add lod tensor array concat
test=develop
* revise some bug of lod tensor array concat
test=develop
* add unittest for tensor array concat
test=develop
* change to tensor array to tensor
test=develop
* revise bug
test=develop
* revise a bug
test=develop
* revise a bug
test=develop
* revise a bug of python3
test=develop
7 years ago
6c2b891d87
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
e23061e0dc
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
6c6e638550
Add InferVarType for some op ( #14201 )
...
* add_infer_var_type
test=develop
* InferVarTypeHelper-> VarTypeInferenceHelper
test=develop
* PassInputTypeAndDTypeOnOutput
test=develop
* follow comment
test=develop
7 years ago
664a4e010c
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
1eec5a428f
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
0b38822624
Merge pull request #14345 from heavengate/fix_grid_sampler
...
fix #14344 : win compile error, EigenTenor * float unsupport. test=develop
7 years ago
6f9c70acb7
Merge branch 'windows/build' into windows/online
...
test=develop
7 years ago
ca60e1d34d
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
433fc7c1d4
skip mkldnn related pass when use_mkldnn=false
...
test=develop
7 years ago
4bd0c4c5ee
test=develop
7 years ago
350f1f3971
remove duplicate function definition
7 years ago
4b1f1a8787
fix merge issue
7 years ago
d08334011a
fix merge issue
7 years ago
6ae0b91b39
Clean LockGuardPtr
...
test=develop
7 years ago
52f7644f53
Merge remote-tracking branch 'upstream/develop' into windows/build
7 years ago
1420c3b155
Add enum AllocatorStrategy
...
test=develop
7 years ago
698698f2fa
Merge branch 'develop' into fix_vlog
7 years ago
abe209234f
Exhaustive search for cuDNN conv. ( #14286 )
...
* exhaustive search for cuDNN conv.
* Refine code and add unit testing.
* Fix model load in fluid/inference and unit testing in conv2d
* Follow comments.
* Fix compiling test=develop
7 years ago
b59a9bfb7c
Clean buffered_allocator
...
test=develop
7 years ago
f215534ecf
Merge pull request #14205 from heavengate/nearest_interp
...
Add interpolate operator replace bilinear_interp_op and add nearest neighbor interp mode
7 years ago
72108d8dbe
fix win compile error: EigenTenor * float unsupport. test=develop
7 years ago
26fb34c365
Merge develop tiny fix
7 years ago
fdc689142c
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rewrite_allocation
...
test=develop
7 years ago
7ffc9fd839
Merge branch 'rewrite_allocation' of https://github.com/sneaxiy/Paddle into rewrite_allocation
7 years ago
22125ebaef
Merge pull request #14321 from tensor-tang/fea/jit/vscal
...
Fea jitcode vscal vaddbias
7 years ago
f1046d7e37
Merge pull request #14335 from wojtuss/wojtuss/add-graph-viz
...
added additional call to graph_viz_pass
7 years ago
34e9e59f4a
Merge pull request #14333 from kbinias/change-hardcoded-format-and-bump-mkldnn-version
...
Changed hardcoded format to any in convolution and bumped MKL-DNN version to 0.17rc
7 years ago
3f91e0f001
update API.spec
...
test=develop
7 years ago
b5f617fa9b
make mobilenet test reuse resnet50 test
7 years ago
1987d45e75
add comment for depthwise pass
7 years ago
87450b9ad4
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_vlog
...
test=develop
7 years ago
41b423d41b
remove duplicate
7 years ago
dcfab11193
merge from develop
7 years ago
4ffa92d4f0
Merge branch 'develop' into windows/build
7 years ago
c5b6573a5a
Fix input<tensor> ( #14208 )
...
* fix input<tensor>
test=develop
* fix split_ids
test=develop
* ElementwiseMul should not support SelectedRows
* fix scale op
test=develop
* change GetTensorFromVar() method to GetTensorOrSelectedRowsFromVar()
* fix operator
* refine MultiOutput
* fix MultiOutput
test=develop
* disable test_dist_save_load
test=develop
* fix elementwise_op
test=develop
* add get_sparse_as_op
test=develop
* add info for check
test=develop
* rename get_sparse_as_op with extract_rows_as_op.
test=develop
* elementwise doesn't support selected_rows
* fix regularizer
* remove extract_rows_as
test=develop
* fix ci
test=develop
* add test for sum_op
* fix regularizer
test=develop
* test=develop
* fix pserver weight decay multi inputs test=develop
7 years ago
f1c1acf1ac
Changed hardcoded format to any in convolution and bumped MKL-DNN version to 0.17-rc
...
test=develop
7 years ago
813e54efbd
Merge pull request #14328 from PaddlePaddle/revert-14046-windows/debug
...
Revert "cherry picked windows patches."
7 years ago
3db9fad764
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_vlog
...
test=develop
7 years ago
3da43dcae2
Because anakin do NOT use glog, so we revert anakin related change
...
test=develop
7 years ago
387610aae1
Merge pull request #14325 from luotao1/fix_test_analysis_predictor
...
fix test_analysis_predictor
7 years ago
45125ba538
fix share library issue
7 years ago
b03a44e062
Merge pull request #14026 from JiabinYang/add_reorg_op
...
Add reorg op
7 years ago
ff6c809bfc
Merge pull request #14251 from panyx0718/fix
...
Make OpHandle/VarHandle and ir::Node works cleaner
7 years ago
ba8b5619a3
Revert "cherry picked windows patches."
7 years ago
49710960ef
Revert tensor_util.cu
...
test=develop
7 years ago
fcc0452c8b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_vlog
...
test=develop
7 years ago
381bea0a16
fix test_analysis_predictor
...
test=develop
7 years ago
0c3227a523
Change the origin VLOG level to 10 times
...
Fix code to support cpplint syntax check
test=develop
7 years ago
5e64244f25
add vaddbias jitcode
...
test=develop
7 years ago
5f7956ae59
Merge remote-tracking branch 'ups/develop' into fea/jit/vscal
7 years ago
59c66532e7
add more logs and comments
...
test=develop
7 years ago
1f4a434302
Merge pull request #14046 from dzhwinter/windows/debug
...
cherry picked windows patches.
7 years ago
869487a2b7
Merge remote-tracking branch 'origin/develop' into windows/build
7 years ago
7fd640b882
added additional call to graph_viz_pass
...
test=develop
7 years ago
3d950a812d
combine jitcode of vscal
7 years ago
03e11f3fc9
add vscal jitcode
7 years ago
5b7a9dd7ac
Merge pull request #13815 from jacquesqiao/optimize-pyreader
...
optimize pyreader
7 years ago
234a1d9248
Merge remote-tracking branch 'origin/develop' into windows/debug
...
test=develop
7 years ago
a270fdf2db
Fix SelectedRowsAdd bug ( #14309 )
...
* fix selected_rows bug
test=develop
* refine cos_sim
test=develop
7 years ago
ce994190ab
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into optimize-pyreader
...
test=develop
7 years ago
2f0a379af7
Merge pull request #14307 from tensor-tang/fix/mac
...
fix mac
7 years ago
b2af213009
Merge pull request #14292 from sneaxiy/delete_buggy_selected_rows_functor
...
Delete buggy selected_rows functor
7 years ago
161ba9c9d1
fix mac
...
test=develop
7 years ago
f395075efc
rebased and stuff broke
7 years ago
e8642c3c1f
Merge pull request #14265 from tensor-tang/fea/jit/vadd
...
add vadd, vaddrelu jitcode
7 years ago
a60957f386
addd test_analyzer_mobilenet
7 years ago
8b47d90f5d
add 'actual_shape' attribute. test=develop
7 years ago
382307b943
refine code
...
test=develop
7 years ago
3319072858
fix jit kernel test on mac
...
test=develop
7 years ago
44cb70c088
Merge remote-tracking branch 'ups/develop' into fix/mac
7 years ago
c774bcbd2d
Merge device_context
...
test=develop
7 years ago
057a682ee9
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rewrite_allocation
7 years ago
c28beb8a3c
test(Pe): add dry run tests for pe ( #14254 )
...
Dry run tests will skip `Op.Run` and just perform job scheduling. It helps to analysis dead lock in PE.
test=develop
7 years ago
c9730d33d9
fix run error on mac
...
test=develop
7 years ago
80132933b7
Merge pull request #14281 from luotao1/face
...
refine analysis_resnet50_tester
7 years ago
e0c8397426
Merge pull request #14257 from jacquesqiao/optimize-pserver-profiler-thread-pool
...
clean rpc server profiler
7 years ago
ffc866159f
hot fix log ( #14293 )
...
test=develop
7 years ago
65b61db10a
Merge pull request #13927 from NHZlX/fix_googlenet_bug_with_rule
...
Fix googlenet bug with rule
7 years ago
25e070ecc7
Merge remote-tracking branch 'ups/develop' into fea/jit/vadd
7 years ago
ef8218be22
update docs test=develop
7 years ago
eea36739cc
refine test_helper.h
...
test=develop
7 years ago
6449faec37
Merge pull request #14259 from jacquesqiao/optimize-thread-pool
...
Optimize thread pool
7 years ago
9518bc8d0a
delete buggy selected_rows functor
...
test=develop
7 years ago
a9b5d42dd4
Add fp16 backward support ( #14202 )
...
* add fp16 backward support
test=develop
* add sum_op fp16 test
* disable test_dist_save_load
test=develop
* add check_grad for sum
* add unit test for softmax_grad fp16
test=develop
* add scale_op unit test
* add mul_grad_op unit test for fp16
* add cross_entropy_grad and eman_grad unit test for fp16
test=develop
* fix cross_entropy unit test
* add pool2d fp16 unit test
* refine conv2d fp16 unit test
test=develop
* refine activation unit test
test=develop
* fix ci
test=develop
* follow zhihong's comment, copy from https://github.com/PaddlePaddle/Paddle/pull/12796
test=develop
7 years ago
3b8dd9ebbd
optimize code test=develop
7 years ago
2b791f1f63
unify analyzer_face_tester to analyzer_resnet50_tester
...
test=develop
7 years ago
2921f8a79c
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into optimize-pserver-profiler-thread-pool
7 years ago
1ead9318d5
remove unused code in test_helper.h to pass ci
...
test=develop
7 years ago
4062f00f2a
optimize thread pool code
...
test=develop
7 years ago
2835e04409
merge develop branch. test=develop
7 years ago
deb4af70ef
add test
7 years ago
db8c52da5e
Revert " Exhaustive search for cuDNN conv. ( #14043 )"
...
This reverts commit ce7d9b0799
7 years ago
ce7d9b0799
Exhaustive search for cuDNN conv. ( #14043 )
...
* exhaustive search for cuDNN conv.
* Refine code and add unit testing.
* Clean code
* Fix model load in fluid/inference and unit testing in conv2d
* Follow comments.
7 years ago
cb4083b9fa
fix compile error
...
test=develop
7 years ago
dd343a4971
Merge remote-tracking branch 'ups/develop' into fea/jit/vadd
7 years ago
Tao Luo
Jiabin Yang
Yu Yang
hjchen2
Qiao Longfei
tensor-tang
peizhilin
Zhaolong Xing
Yu Yang
Dang Qingqing
jerrywgz
chengduo
Yiqun Liu
Yihua Xu
Yihua Xu
Superjomn
dengkaipeng
Wu Yi
Jacek Czaja
Michal Gallus
Tomasz Patejko
whs
Qiyang Min
minqiyang
Sylwester Fraczek
dzhwinter
Xin Pan
Zeng Jinle
baojun-nervana
sneaxiy
Yibing Liu
ruri
li099
Krzysztof Binias
Wojciech Uss
dzhwinter
barrierye