d41176411f
Update test_py_reader_using_executor.py
8 years ago
b084dfab7e
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into parallel_bcast
8 years ago
5ce1a960a5
move bcast op into pass
8 years ago
36d6e44681
fix test_py_reader_using_executor error
8 years ago
392ae69650
Set parallel executor thread num under nccl2 distributed env ( #13207 )
8 years ago
0c1a5d87b4
"debug version"
8 years ago
dfbd1cc3c1
Merge pull request #13209 from Sand3r-/mgallus/conv-relu-fuse
...
[MKLDNN] Fuse Conv+BatchNorm + ReLU
8 years ago
5d34ef61cb
Fuse MKLDNN's Conv + ReLU
8 years ago
b681537e1a
Add multiprocess reader ( #13311 )
...
* add multiprocess_reader
* add multiprocess_reader to reader decorator
* support piped multi process reader
* revert v2 decorator
* add comment to multiprocess_reader
* optimize code
* use ujson to speed up json serialize/deserialize
* add assert to multiprocess_reader
* update comment of multiprocess_reader
* optimize ujson import, handle error case
* optimize import ujson
* remove ujson from requirements.txt
* add import sys to decorator.py
8 years ago
cc18fffb90
add nest while_op
8 years ago
faf8ad2436
Add ignore_index in cross_entropy op ( #13217 )
...
* add ignore index
* update api.spec
* enhance softmax_with_cross_entropy
8 years ago
17bf8713a5
Merge pull request #12988 from panyx0718/ir2
...
program and tensor versioning support
8 years ago
314e150f8a
Merge branch 'public-dev' into develop
8 years ago
a39eba77eb
Implement norm_op by CUDA instead of Eigen. ( #13273 )
...
* Implement norm_op by CUDA instead of Eigen.
* Remove the commented code.
8 years ago
fbcecddfce
Merge branch 'public-dev' into develop
8 years ago
182b24ce3c
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into quantize_transpiler_update
8 years ago
dc863aac7e
Add kids exists detection in Scope
8 years ago
9b7c3f9615
refine
8 years ago
56a977d436
add test
8 years ago
926e1077ca
version
8 years ago
f90c7865f0
Benchmark tool for imgnet ( #12305 )
...
* support test using executor without reader
* run imgnet
* update fluid benchmark
* wip
* update
* update all models
* support pyreader
* update
* clean up
* make profile batches contollable
* update API.spec
* update scripts
* clean dockerfile
* update
* clean comments
* add scope argument docstring
* use num_trainers to determine nccl init comms
8 years ago
5e7bfd4e4c
Merge branch 'develop' into anakin_ner
8 years ago
ce24a92007
Disable image_classification_resnet
8 years ago
d3d22a12fd
Polish code
8 years ago
6fae46a128
refine codes
8 years ago
5335ff628e
Polish code
8 years ago
2720330dc5
Disable random fail case
8 years ago
8ef1f9f9b8
Polish code
8 years ago
2bb0ac927b
Polish code
8 years ago
a04d99810e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_python35_CI_random_fail
8 years ago
9f2ccf5b3d
Merge pull request #13237 from tensor-tang/refine/op/peephole
...
refine fusion lstm/peephole and fusion gru
8 years ago
a121c89830
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into quantize_transpiler_update
8 years ago
6a79ba2f28
Add save inference model to trainer ( #12682 )
...
* add save_inference_model for Trainer
* add comment
* update comment
* fix mac compile
* add unit test
* update API.spec
* revert cpplint-cpp-source
8 years ago
ad9e6476ff
Force object deletion on trainer in unit test
8 years ago
718033e14f
Merge remote-tracking branch 'ups/develop' into refine/op/peephole
8 years ago
5af0c60fb6
Add compat deps
8 years ago
ad4d965172
Merge pull request #13250 from Yancey1989/fix_dist_base
...
fix parallel run dist unit test
8 years ago
e27a1a6a6a
Merge pull request #13229 from chengduoZH/disable_feed_fetch_unit_test
...
Disable feed_fetch_unit_test temporarily
8 years ago
40dbd97f8e
Merge remote-tracking branch 'ups/develop' into refine/op/peephole
8 years ago
3420c94994
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_python35_CI_random_fail
8 years ago
d4accfa905
Fix random fail of python35 unit test
8 years ago
b805751598
Merge pull request #13223 from velconia/open_python35_CI
...
Open python35 ci
8 years ago
0a71d58089
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into fix_dist_base
8 years ago
a0b68653d0
make pickel proto version compatiable with py2
8 years ago
886852557f
Refine reshape_grad and transpose_grad ( #13074 )
...
* Add intermediate
* fix flatten/squeeze/unsqueeze
* Considering compatibility issues, we could not fix the origin op
* follow comment
* reset the shape of XShape
8 years ago
e3c7348f52
Update quantize_transpiler
8 years ago
a267155006
fix parallel run dist unit test
8 years ago
3eb55f0643
Merge remote-tracking branch 'ups/develop' into refine/op/peephole
8 years ago
d7ac1cc836
refine seq when bs is large
8 years ago
6e03f7900f
Add centered mode rmsprop ( #13161 )
...
* rmsprop optimizer support v1 mode
* typo
* optimize code
* refine code
* optimize unit test
* update test_rmsprop_op.py
* update formula of rmsprop
* optimize document
* update API.spec for RMSPropOptimizer
* add default value to check_output_with_place equal_nan
8 years ago
2f3b498949
refine fusion seq lstm peephole
8 years ago
d1e2efae6b
reimplement auc in fluid ( #13167 )
...
* reimplement auc in pyton
* reimplement auc in fluid
* add auc unittest
* replace new auc in layers
* add batch Auc in Fluid
* name formated
8 years ago
73eba2f87e
disable feed_fetch_unit_test
8 years ago
04272c0d41
Enable lstm peephole ( #13160 )
...
* Refine fusion lstm op code for better readability
* Enable peephole in fusion lstm op (seq_mode part) and add unit test
* Enable peephole in fused lstop op (batch_mode part)
Set batch_mode as default as well
* Use pre-commit to clean format
* Follow up review comments as well as adding more unit tests for seq mode
8 years ago
1d0353f162
Quantize transpiler
8 years ago
bab341213e
Remove check in top_k API. ( #13178 )
8 years ago
a0e6f416d6
Port transformer
8 years ago
c838fa3173
Port dist_transpiler to Python3.5
...
Resume prelu_op_test in python2
8 years ago
cda7842e26
Revert "Revert "Add Python Callstacks when Op::Run error ( #12759 )""
...
This reverts commit 1f270275a6
8 years ago
9557cc218d
Refine and fix some code for faster-rcnn. ( #13135 )
...
* Fix bug in generate_proposals_op.
* Fix data type for RoIs.
* Refine and fix rpn_target_assign_op.
* Add the missing file bbox_util.h
* Rename BoxEncoder to BoxToDelta
8 years ago
d9c3123aff
Merge pull request #13181 from tensor-tang/refine/fusion/ut
...
refine fusion lstm and gru op test
8 years ago
6bc16ff33b
Merge pull request #13186 from seiriosPlus/selected_rows_size_fix
...
bug fix
8 years ago
Chen Weihang
3b6090e80b
Merge pull request #12887 from chenwhql/sequence_enumerate_op
...
Feat: add sequence enumerate op
8 years ago
ff618a2880
bug fix
8 years ago
ef71b8e2ae
refine fusion lstm and gru op test
8 years ago
ef628ab887
Fix high level API(Inference) bug ( #13159 )
...
* fix high level API(Inference) bug
* patch the unit tests
8 years ago
fbdd4f8c0f
Merge pull request #13101 from zenghsh3/develop
...
Fix bug of sampling_id op
8 years ago
9bd933d3fb
Improve and fix fake_quantize_op ( #13092 )
...
* Improve and fix fake_quantize_op.
8 years ago
chenweihang
7ddbbcb0b5
doc: refine API and doc
8 years ago
91422138d2
Merge pull request #13104 from jacquesqiao/fix-sparse-grad-merge
...
fix sparse grad merge on pserver
8 years ago
chenweihang
b081363bae
Merge branch 'sequence_enumerate_op' of https://github.com/chenwhql/Paddle into sequence_enumerate_op
8 years ago
chenweihang
0b7d82befb
doc: refine English description
8 years ago
d0c01d319a
fix codestyle
8 years ago
b2de658916
refine code
8 years ago
a64200f8c1
refine unittest of sampling_id op
8 years ago
e10aa80f03
Add pad2d op. ( #12950 )
...
* Add pad2d op.
* Add unitest and python api.
* Fix cuda op kernel.
* Fix python api.
* Fix python api.
* Update API.spec.
* Fix python api
8 years ago
dee6d1606f
Enable conv and batch norm by default
8 years ago
ed511fe5a7
update test_dist_transpiler
8 years ago
f1554a4a07
fix sparse grad merge on pserver
8 years ago
c709a04ae2
Merge pull request #13018 from panyx0718/name
...
support op_namescope for better debugging
8 years ago
0b8067c0dc
fix dist train reduce mode ( #13068 )
...
* fix dist train reduce mode
* fix previous fix
8 years ago
1a88baae22
add rpn_target_assign api test ( #13013 )
...
* Add unit test for rpn_target_assign API.
8 years ago
03b1e4bec9
Merge pull request #13047 from JiayiFeng/dev_mem_analyse
...
refine memory usage calc
8 years ago
0a97d24b41
Faster RCNN Generate Proposal Labels ( #12616 )
...
* Add generate_proposal_labels for Faster-RCNN.
8 years ago
f88a8ba99a
Merge pull request #12793 from kuke/wrap_squeezes
...
Wrap unsqueeze & squeeze ops
8 years ago
3bd1d22a7d
Enhance fused_elementwise_activation_op ( #12837 )
...
* Enhance the function of fused_elementwise_activation_op
* enhance unit test
* Clean Code And Add Doc
* Add compound functors
* Fix doc and enhance unit test
* define Dx and Dy for d_binary_func
* add mul_scale
* add mul_scale
* add elementwise_mul
* code refine
* code refine
* add doc
* add AsIntermediate
8 years ago
a615ad46e4
Add test for dist and memopt ( #13049 )
...
* add test for dist and memopt
* update transformer too
8 years ago
515a756a3c
Merge pull request #13048 from TomorrowIsAnOtherDay/develop
...
add python api for sampling_id op
8 years ago
f9e20747fc
Modify program_to_desc functions to add types support. ( #12991 )
8 years ago
d2d082db19
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_mem_analyse
8 years ago
66f870433d
fix test
8 years ago
51ef0ad766
allow to use name_scope for debugging and visiualization
8 years ago
Chen Weihang
b98b744067
Merge branch 'develop' into sequence_enumerate_op
8 years ago
9ae55dd70d
fix dist transpile with memopt ( #12974 )
...
* fix dist transpile with memopt
* update api.spec
* polish dist transpile api
* update apispec
* update apispec
8 years ago
bbcf1ad263
update
8 years ago
e1f17a1e4f
add python op for sampling_id
8 years ago
fb08e163cf
refine memory usage calc
8 years ago
2ad5d91ef8
Faster RCNN Generate Proposals ( #12056 )
...
* Add proposals generation operator for Faster-RCNN.
8 years ago
01e96a8e20
Add Transformerunittest ( #12968 )
8 years ago
89d6d69ce4
Merge pull request #12781 from tensor-tang/feature/op/fusion_gru
...
add fusion gru
8 years ago
82671e9486
Fix bug in flowers dataset. ( #13024 )
8 years ago
2bb15f437c
Merge pull request #12791 from panyx0718/ir3
...
graph to program pass
8 years ago
8965cee89f
Polish PrintOp ( #12895 )
...
* Polish PrintOp
* Polish PrintOp
* Polish PrintOp
* Refine test_print_op
8 years ago
13509da69e
Merge upstream to branch wrap_squeezes
8 years ago
9be39bb4b7
Enhence optimizer. ( #13004 )
8 years ago
7ad39c4077
Enhance pad_constant_like_op ( #12999 )
...
* enhance pad_constant_like_op
* add API
* add API
8 years ago
0353eddb51
Improve fake_dequantize_op. ( #12877 )
...
* Improve fake_dequantize_op.
* Follow comments.
8 years ago
11e01d9b2d
Scale support selectedrows ( #12960 )
...
* add ScaleOpVarTypeInference for scale op
* scale op support scale selected rows
* optimize code
* use FindVar
* use FindVarRecursive in ScaleOpVarTypeInference
8 years ago
7b84c580e2
Merge pull request #12824 from JiayiFeng/dev_sequence_padding_op
...
Sequence pad op
8 years ago
0ee6fed05b
Refine dist rpc deps ( #12899 )
...
* refine dist train RPC deps
* clean up
* clean up
* fix ut
* remove input for fetch_barrier
* follow comments
8 years ago
7e0c9f50ae
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_sequence_padding_op
8 years ago
599a32641b
Merge pull request #12971 from sneaxiy/unstack_op
...
Add unstack op
8 years ago
1d9cb93276
Merge pull request #12989 from velconia/fix_last_of_py3
...
Fix random diff between python2 and python3
8 years ago
a481c5e98c
Merge remote-tracking branch 'ups/develop' into feature/op/fusion_expand_concat_fc
8 years ago
49c31febb5
fix typo and op test
8 years ago
9cb455fa7d
update function
8 years ago
312f3b8654
Fix random diff between python2 and python3
8 years ago
52a480bb98
Merge develop
8 years ago
02909335e9
rename fusion seq_concat_fc to fusion seqexpand_concat_fc
8 years ago
0f0d48230c
add fusion seq_concat_fc op test
8 years ago
1a67061fee
graph to program pass
...
fix a few other things
8 years ago
1f09bc320c
Support data type int8_t . ( #12841 )
...
* Support int8 type.
8 years ago
chenweihang
0c4697f8cd
fix: change to enumerate by sentence
8 years ago
24264bc0b8
Merge develop
8 years ago
0153c21d83
add unstack_op
8 years ago
3e1050a2e8
Add pad_constant_like_op ( #12943 )
...
* Add pad_constant_batch_size_like
* refine pad_op
* optimize memory
8 years ago
c488ee96a7
Merge remote-tracking branch 'ups/develop' into refine/op/fusion_lstm
8 years ago
e61cf3214d
complete reverse seq
8 years ago
Chen Weihang
4ec12496dd
Merge branch 'develop' into sequence_enumerate_op
8 years ago
1777cd09f6
refine fusion lstm op test
8 years ago
4b28fab8c9
enable more acts
8 years ago
3c58b87b45
fix auc layer and add check for auc op ( #12954 )
...
* fix auc layer and add check for auc op
* use input to check if states are inited
* optimize code
8 years ago
efa6f0df4e
Remove inplace args in unsqueeze & squeeze ops
8 years ago
5aa64827df
Rename the input of squeeze's wrapper
8 years ago
b586cc2ae8
Fix typos in unsqueeze & unsequeeze wrapper
8 years ago
1443d762fd
Wrap unsqueeze & squeeze ops
8 years ago
d189d4dbab
Merge pull request #12884 from sneaxiy/sequence_mask_op
...
Add sequence_mask_op for DAM model
8 years ago
7240e7215a
Merge pull request #12914 from JiabinYang/check_cv2
...
fix bug to avoid warning once import paddle.fluid
8 years ago
038e2817f0
Merge pull request #12836 from velconia/disable_prelu_test_local
...
Disable prelu_op_test until fixing Python3 issues
8 years ago
ef6445ee39
Merge pull request #12908 from seiriosPlus/fill_constant_selectedrows
...
add SelectedRows support in fill_constant_op
8 years ago
786558fc68
fix bug to avoid warning once import paddle.fluid
8 years ago
chenweihang
733ea0d29b
adjust infershape details
8 years ago
836e1e0b36
Merge pull request #12910 from velconia/fix_flowers_dataset
...
Fix flowers dataset download problem
8 years ago
66cc1850a8
add gpu place
8 years ago
fff6f595ff
add unit test
8 years ago
8f9bbc2834
add unit test
8 years ago
fca139b5e3
Fix flowers dataset download problem
8 years ago
557be6fc58
Merge pull request #12902 from PaddlePaddle/revert-12736
...
Revert "Disable in_place in batch_norm API. (#12736 )"
8 years ago
a7849db561
Port new added code
8 years ago
6c775a0a06
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into disable_prelu_test_local
8 years ago
Chen Weihang
2969aba14f
Merge branch 'develop' into sequence_enumerate_op
8 years ago
chenweihang
219a2369da
feat: wrap sequence enumerate op
8 years ago
7df74a5ee4
Merge develop
8 years ago
1f270275a6
Revert "Add Python Callstacks when Op::Run error ( #12759 )"
...
This reverts commit b2df17003f
8 years ago
b1fc238694
Revert "Disable in_place in batch_norm API. ( #12736 )"
...
This reverts commit f5d5d7b2d9
8 years ago
1083e99520
Merge develop
8 years ago
7570e5ef04
Print readable program codes. ( #12673 )
8 years ago
f0f06992c1
Merge pull request #12878 from tensor-tang/feature/op/attention_lstm
...
Add attention lstm cpu forward
8 years ago
5ea7bf88ba
Merge pull request #12872 from sneaxiy/stack_op
...
Add stack_op for DAM model
8 years ago
e895c98f0a
add support to max_len is None
8 years ago
0fb5e351c1
update API.spec
8 years ago
709c37023a
Polish code
8 years ago
f4a4a4cbd9
add op comment and python layer
8 years ago
chenweihang
d2e5395b97
feat: add sequence enumerate op
8 years ago
8ad9055804
Add is_test for while_op ( #12874 )
...
* add is_test for while_op
* Change API
8 years ago
80e3ce411d
For test
8 years ago
b8da70c370
Resovle multi gpu async deps ( #12828 )
...
* dist transpiler add control dependency var between send and recv
* fix async deps
* follow comments and refine
* fix deps connect for rpc ops
8 years ago
64464cb1fa
Merge develop
8 years ago
79918a8442
add sequence_mask_op for DAM model
8 years ago
b2df17003f
Add Python Callstacks when Op::Run error ( #12759 )
...
* Add Python Callstacks when Op::Run error
* Skip op with sub-block
* refactor: refine callstack info's format
* Reshape only support matrix
* Polish Python code
* Fix UT
* Fix Py3
8 years ago
c95ff1c410
Add debug info
8 years ago
5ca0bb9aad
support more activation type and remove some comments
8 years ago
31f587e7e0
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into disable_prelu_test_local
8 years ago
01eec0af91
Fix flowers dataset reading problem
8 years ago
d9bf73f3ab
Merge remote-tracking branch 'ups/develop' into feature/op/fusion_gru
8 years ago
dd938d0b94
fix bugs and pass op test
8 years ago
522b3e411f
complete attention lstm op test
8 years ago
93cc29abc0
init attention lstm op test
8 years ago
064b7f3de1
Change the md5sum of 102flowers dataset
8 years ago
d49a0d755b
Fix common download problem
8 years ago
f5d5d7b2d9
Disable in_place in batch_norm API. ( #12736 )
...
* Disable in_place in batch_norm API.
8 years ago
c73c5ed573
use for_range
8 years ago
6d107b0f39
Fix the test_desc_clone's problem
8 years ago
7748963479
refine op_test ( #12846 )
8 years ago
b548ecbc2b
add stack_op
8 years ago
e8b4e0d627
fix load_vars bug ( #12869 )
8 years ago
6d9b9cb4b6
Add debug info for anakin cpu
8 years ago
57dab0bb4c
Change the link of flowers
8 years ago
94f6e54db9
Add timeout for python3
8 years ago
bcaa1d5e73
Merge pull request #12853 from panyx0718/profiler
...
fix profiler test
8 years ago
f8c6b46415
fix profiler test
8 years ago
6fad27fc64
add fusion gru op test
8 years ago
5424a698df
refine gru op unit test
8 years ago
39c526d42f
Port test_dist_transpiler to it
8 years ago
efe88ab9f9
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into disable_prelu_test_local
8 years ago
cd32ddac12
Fuse Convolution and Eltwise Add into MKLDNN's Conv+Bias ( #12669 )
...
* Fuse Convolution and Eltwise Add into Conv+Bias
* Reduce bias branching at conv_mkldnn_op
* Add MKLDNN build checks for Conv Bias
* Conv-bias: check if bias input exist befor assignment
* Conv-bias: Remove Bias dim check from infershape
It was causing conv3d test to crash upon\ncalling HasInput(Bias)
8 years ago
d94a3f621b
Disable prelu_op_test until fixing Python3 issues
8 years ago
cbc6e6eb97
Merge pull request #12247 from seiriosPlus/dis_ckpt_fix
...
add load slice_vars in io.py
8 years ago
72965226e6
Merge pull request #12818 from velconia/fix_python3_CI_job
...
Fix python3 CI job
8 years ago
f63368db5e
Add async dist tests ( #12798 )
...
* add async dist tests
* update delta
* fix transformer test
* refine rmsprop transpile
* update
* fix dist seresnet
8 years ago
9f33222251
Merge pull request #12822 from luotao1/debug_no_patchelf
...
only change rpath in Release mode
8 years ago
08152916cf
add annotation in load_inference_model
8 years ago
3d11d018e0
Fix scatter_op python API ( #12742 )
...
* Fix scatter_op python API and remove inconsistency between implementation and doc
* API spec change
* Change as review comment
8 years ago
2049139f9f
only change rpath in Release mode
8 years ago
5a83776751
Port test_desc_clone
8 years ago
50d66a0790
Fix prelu_op
8 years ago
7dbf5f1244
Update distribute_transpiler.py
8 years ago
c0a95bb2d8
Update io.py
8 years ago
3ae97aab6c
fix trainer load parameters ( #12816 )
8 years ago
beb93bb901
Fix ut bug for graph_test
...
Port dist_transpiler new added codes
Port ut for clone desc
8 years ago
e955361267
Merge pull request #12737 from tensor-tang/feature/op/fusion_lstm
...
add fusion lstm
8 years ago
Chen Weihang
57b34d9196
Merge pull request #12808 from chenwhql/remove_inplace_param_in_squeeze_and_unsqueeze
...
Refactor: remove inplace parameter from squeeze and unsqueeze op
8 years ago
c62f68cb94
Fix bug in conditional_block_op. ( #12246 )
...
* Fix bug in conditional_block_op.
* Fix bug and add comments.
* Rename arguments.
8 years ago
chenweihang
bc471b6ac4
refactor: remove inplace parameter from squeeze and unsqueeze op
8 years ago
c6af7201e9
Merge pull request #12692 from reyoung/feature/fast_executor
...
Feature/fast executor
8 years ago
5fa1f05c9d
code clean and rename
8 years ago
bf2f1599f9
code clean
8 years ago
674b719bf4
add print
8 years ago
b090479409
Merge remote-tracking branch 'ups/develop' into feature/op/fusion_lstm
8 years ago
f6d073422c
refine fusion lstm op test
8 years ago
3efac174ea
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into sum_op_dim_fix
8 years ago
8cfcbabf5e
fix dist unittest
8 years ago
dbb4f0d35d
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into dis_ckpt_fix
8 years ago
fd10669ecb
Add dependency to send recv ( #12760 )
...
Add dependency to send recv
8 years ago
8d8d48a34f
Complete sequence_pad_op and its CPU kernel. Add unittests
8 years ago
f4999e4f39
Remove unused six and missing six import
8 years ago
c24a9263ba
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into sum_op_dim_fix
8 years ago
7715cdc69b
add unit test in dist transpiler
8 years ago
bb9f494740
merge develop
8 years ago
eba7177475
add unit test and code fix
8 years ago
653fad08f8
Optimize selected rows for dist lookup table with pthread rwlock ( #12635 )
...
Optimize selected rows for dist lookup table with rwlock
8 years ago
479a443f68
name optimized
8 years ago
0749c8822d
Merge pull request #12556 from seiriosPlus/samplingIdOp
...
Sampling id op
8 years ago
1ef5f2c3e8
Make flowers reader and parallel_executor more efficient
8 years ago
e398348e7e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
8 years ago
340a104c58
Merge pull request #12658 from velconia/port_pybind11
...
Port pybind11 and python code to support py3 CI test
8 years ago
125e9166e1
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into sum_op_dim_fix
8 years ago
b599bd8d46
Merge remote-tracking branch 'ups/develop' into feature/op/fusion_lstm
8 years ago
d7873e1412
remove patchelf in windows ( #12710 )
...
* remove patchelf in windowls
* "follow comment"
8 years ago
a72f68f223
Merge remote-tracking branch 'ups/develop' into feature/op/fusion_lstm
8 years ago
df28a3b452
fix lod and op test
8 years ago
f3cd2612ae
refine fc and use the fc compute in fusion_lstm
8 years ago
c44fb00371
Add name in relu and log API. ( #12438 )
8 years ago
40138c4cd6
add unit test of fusion lstm op
8 years ago
99d3f08920
Add print_function for all python files
8 years ago
c108376506
Add three modes for prelu_op ( #12630 )
...
* Add three modes for prelu_op.
8 years ago
d06849305a
parameter dispather. ( #12666 )
8 years ago
efc5392d97
Merge pull request #12676 from tensor-tang/refine/op/fc
...
refine fc op
8 years ago
a32ce8c444
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
8 years ago
baa6273c54
unit test optimize
8 years ago
507f479732
Polish code
8 years ago
09e30eeeda
Merge pull request #12696 from seiriosPlus/lookuptable_size_on_pserver_fix
...
lookup table size fix
8 years ago
0d7047ca79
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
8 years ago
d1d825ee02
Hide unnecessary API
8 years ago
265302edea
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/fast_executor
8 years ago
13e99cf92f
add unit test
8 years ago
bd87f67f0e
Dist transpile can pass startup program by argument ( #12606 )
...
* dist transpile can pass startup program by argument
* update API.spec
8 years ago
9333a62792
Add flatten op interface and enhance APIs about detection to support variable-length image. ( #12422 )
...
* add flatten api&enhance detection api
* unify shape_op data type
* update API.spec
8 years ago
ed937bc6f8
merge
8 years ago
e73fc973d6
revert trainer py
8 years ago
4d4491ef6a
Fix new added code
8 years ago
77f12e000f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
8 years ago
d6b5302bd6
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_op_tensor_support
8 years ago
0098a494a2
Merge remote-tracking branch 'ups/develop' into refine/op/fc
8 years ago
842fb021b3
Fix clone() bug. ( #12583 )
8 years ago
5d579e1a96
add export_for_deployment flag to save_inference_model ( #12582 )
...
add export_for_deployment flag to save_inference_model
8 years ago
1f86c88f4a
Remove random order of fetch_list in test_random_crop_op
8 years ago
f42247e55f
change to int to compatible with py3
8 years ago
e0d5f8a820
Move compat module to python/paddle
8 years ago
0707abb51b
lookup table fix
8 years ago
5338417b47
Polish code style
8 years ago
ae39709e59
Polish code
8 years ago
1800fef142
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
8 years ago
05cadf1b24
Add FastExecutor
8 years ago
b9dbb7c5cb
fix bias attri in mkldnn fc
8 years ago
64683cf84a
fix
8 years ago
011ff6bce1
bug fix
8 years ago
45d0259a67
add fc forward test
8 years ago
59580a7f69
bug fix
8 years ago
8b77448d5f
hide misc APIs ( #12540 )
...
* hide misc APIs
* update
* fix transformer test
* update API.spec
8 years ago
d3da0ef9f6
Fix dist train with rmsprop ( #12649 )
...
* fix dist train with rmsprop
* add rmsprop transpiler test
* update by comment
8 years ago
989cae25dd
fix the numpy version has to be lower than 1.15 ( #12523 )
...
* fix the numpy version has to be lower than 1.15
* Add TODO comments
8 years ago
92aa20616d
Polish the code style
8 years ago
68b221401d
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
8 years ago
3972ba3272
add ditriubted attrs
8 years ago
af8b1b993f
Fix rpn_target_assign API.
8 years ago
812de6e806
Port utils to Python3
8 years ago
051eaa5fc7
add ditriubted attrs
8 years ago
057a64508c
enable lookup table to inference
8 years ago
f1a10ccea9
enable lookup table to inference
8 years ago
49ad570ee8
Merge pull request #12667 from luotao1/wmt14
...
fix wmt14 download error
8 years ago
649f5d74f0
fix mine_hard_example bug ( #12664 )
8 years ago
f54efd027e
Merge pull request #12660 from typhoonzero/polish_dist_unittests
...
polish dist unitest
8 years ago
5c537941c2
merge develop
8 years ago
6988aea989
Fix long type in Python3
8 years ago
fae5c1f514
Fix the input of executor
8 years ago
d54267094e
fix wmt14 download error
8 years ago
5b452bfd59
Remove the overfix of lod_rank_table
8 years ago
47561c34b0
Fix python2 CI issues
8 years ago
069ff14756
polish dist unitest
8 years ago
a3539845f2
Polish python code style
8 years ago
e7c7cbaa42
Port new added files to Python3
8 years ago
331151f065
Merge pull request #12647 from jacquesqiao/add-RPCServerProfiler
...
add RPCServerProfiler, replace listen and serv optimizer
8 years ago
e4057d071b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
8 years ago
0b81801515
change init flags
8 years ago
3f2aa91970
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into timeline-support-pure-cpu
8 years ago
64a4925cb4
Merge branch 'Pdv' into samplingIdOp
8 years ago
29fac3c092
Merge pull request #12390 from velconia/port_python3_syntax
...
Apply 2to3 to current paddle main python code
8 years ago
5a9ae411e0
Merge pull request #12618 from sfraczek/sfraczek/fix-new-mkldnn-conv-tests
...
fix UT for mkldnn 0.15
8 years ago
23447cd5a0
Fix parallel_executor_fetch_feed issue
8 years ago
d080d3e694
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into timeline-support-pure-cpu
8 years ago
5d4238cdcc
Fix six.iteritems problem
8 years ago
e4e9450e88
Fix random crop op problem
8 years ago
39059eb70f
Polish code
8 years ago
c2a437fd0f
review fixes
8 years ago
dcbc4284aa
conv2d tests: stop testing on CUDA when use_cudnn=False. Especially when use_mkldnn=True.
8 years ago
d4b10eef5f
Polish code
8 years ago
76ee482e18
Fix cv2 issues
8 years ago
772ceee395
Merge pull request #12615 from typhoonzero/move_dataset_download_url
...
move download urls
8 years ago
6dc07e7f95
Replace items() with six.moves.iteritems() to improve memory usage
8 years ago
bc12c2c616
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_python3_syntax
8 years ago
9cd5999032
Fix dist transpiler unordered dict issue
8 years ago
445ca3db94
Merge pull request #12607 from jacquesqiao/add-unit-test-for-async-transpile
...
Add unit test for async transpile
8 years ago
be6ecec46f
Fix unittests' division issues
8 years ago
59adf7ced1
Fix round(0.0) special issue
8 years ago
7c649e06c3
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into timeline-support-pure-cpu
8 years ago
2ae32f0bf0
revert the change of api
8 years ago
938a9e4faa
Polish code
8 years ago
09103084d3
Polish compat.py and add unittest for it
8 years ago
d74bb6ab9c
fix ut for mkldnn 0.15 - added forcing layout NCHW in mkldnn conv tests
8 years ago
0954adbdcd
move download urls
8 years ago
c3fdf3aee4
Fix divide problem in CI
...
Fix pb_protobuf2 FromString problem
8 years ago
1b6902133d
add TestAsyncLocalLookupTable
8 years ago
8da651057c
add TestAsyncDistLookupTable
8 years ago
24d51de022
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_op_tensor_support
8 years ago
27df3a9f2b
make cross_entropy_op supporting tensors
8 years ago
Chen Weihang
b2435a3a11
Merge pull request #12374 from chenwhql/py_calc_memory
...
Add memory usage estimate API
8 years ago
db7d8136a3
Fix CI issue
8 years ago
ee1d08abba
Fix CI issues
8 years ago
46a2694633
Fix exception problem
8 years ago
c2fce7dd24
Fix bos problem
8 years ago
2a799e3bf2
Fix dist_se_resnet problem
8 years ago
e102e5dd83
Fix cifar10 decompress problem
8 years ago
3ec6d60c34
Fix write bytes in dataset download
8 years ago
e6ae1e4ffc
Replace the dependency of paddle.v2 dataset
8 years ago
7c8b69c700
Feature/op fusion ( #12240 )
...
* Add Preface
* Add demo code
* Save file
* Refine code
* seems can work
* use elementwise strategy
* Use ElementwiseComputeEx
* Add comments
* extract functions from operator
* Refine code
* Follow comment
* code refine
* follow comments
* follow comments
8 years ago
6abe819f07
Fix pybind11 problem
...
Fix str and bytes problem
Fix sorted problem
Fix math problem
Fix CI problem
8 years ago
7970ab965a
Merge pull request #12544 from jacquesqiao/dist-lookup-table-only-support-sgd
...
dist lookup table only support sgd
8 years ago
2d036c47cd
polish dist unit test code ( #12512 )
...
* polish dist se resnext ut
* update
* update
* update
* avoid cpu initializer differ
* change to use executor for now
* update by comment
* remove lr decay use para exe, should fix para exe bug later
* update by comment
8 years ago
97a77512b4
Fix the order of sum ( #12562 )
...
* fix the order of sum
* add doc
* check whether need to copy
* follow comments
8 years ago
111bde9725
add warning to dist lookup table optimizer
8 years ago
b2f28878e1
remove unused code
8 years ago
5a608fcf08
add TestDistLookupTable
8 years ago
c0e8dd8758
add unit test for dist lookup table
8 years ago
3300a532aa
Merge pull request #12546 from reyoung/feature/fix_nmt_cannot_save_checkpoint
...
Do not set loss@Grad as persistable
8 years ago
fd53fdf84b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dist-lookup-table-only-support-sgd
8 years ago
23252a3432
Merge pull request #12463 from jacquesqiao/remove-duplicated-init-op
...
avoid duplicated init op for one parameter
8 years ago
93152b0bac
Fix the result of unit test ( #12520 )
...
* fix the result of unit test
* fix resnext
* compare the result of PE and Exe
* compare the result of reduce and allreduce
8 years ago
5bfdefae91
Merge branch 'Pdv' into samplingIdOp
8 years ago
641a7fb342
fix test_initializer
8 years ago
4cd504d3b4
bug fix
8 years ago
937061a4aa
update test_initializer.py
8 years ago
8aeac78282
revert initializer.py
8 years ago
aa546389e8
move _is_inited_by into framework.py
8 years ago
da2cc99f67
sampling op optimize
8 years ago
1f618c4ff9
Fix the overfix of 2to3 in xrange
8 years ago
55ff03b7ab
Do not set loss@Grad as persistable
...
Revert part of a3ca4c99eb
8 years ago
a7448d5426
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into remove-duplicated-init-op
8 years ago
b656d97e86
Merge pull request #12485 from JiayiFeng/dev_ops_tensor_support
...
Make lookup_table_op and softmax_op supporting high rank tensor
8 years ago
3cea440060
enforce test_parallel_executor_transformer and test_image_classification_resnet to run in serial ( #12545 )
8 years ago
fd73f2bb26
dist lookup table only support sgd while other dense table support all optimizer
8 years ago
9a7084c1ae
fix test_dist_transpiler
8 years ago
28c4443c22
add test for remove duplicated init op
8 years ago
8e9f338c0f
add _is_duplicated_init_op
8 years ago
37e069eba3
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into remove-duplicated-init-op
8 years ago
5656f64b58
Remove duplicate handle of dist_transformer
8 years ago
a58dd3e557
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_python3_syntax
8 years ago
850c3945f2
Make transformer unittest serial
8 years ago
c3e187896d
Port contrib to python3
8 years ago
cd27e53667
Fix python code in contrib
8 years ago
Chen Weihang
3185d38af4
Merge pull request #12467 from chenwhql/batch_size_check
...
fix: add batch size check in loading data process
8 years ago
766ac488ac
sum_op selectedRows dim bug fix
8 years ago
900d61dd98
Clean python api ( #12406 )
...
* api clean
* update API.spec
8 years ago
0c8fde7dce
"cherry picked cpp tests" ( #12182 )
...
* "cherry picked cpp tests"
* "cherry picked"
* "cherry picked tests"
* "merge develop branch"
8 years ago
595a2c83ae
explicit gradient of elementwise_add/elementwise_sub ( #11970 )
...
* "add gradient register"
* "make some enhance"
* "better format"
* "fix typo"
* "fix reuse"
* "fix get expected kernel"
* "change the mkldnn code"
* "fix mkldnn"
* "fix mkldnn failed test"
* "add comment"
8 years ago
23aebf0ea7
update softmax layer comment
8 years ago
chenweihang
bcc2fbe687
add type conversion for batch_size
8 years ago
c237795100
Reopen test_parallel_executor_transformer ( #12452 )
...
* reopen test_parallel_executor_transformer
* save the CPU runing time
8 years ago
chenweihang
8627ef3330
refactor: simplify unittest function
8 years ago
b5a3e40bb5
close random fail test ( #12482 )
8 years ago
c58af84c66
Merge pull request #12426 from panyx0718/better_dist_test
...
Add transformer dist test and factor out some common utils.
8 years ago
7a495a5871
add threads in thread pool for CPU ( #12475 )
8 years ago
f7bd0b227b
Add unittests for softmax_op
8 years ago
12ea358c06
polish
8 years ago
chenweihang
964d631cfd
feat: add unittest of memory usage
8 years ago
b1af7e5d9b
Add unittests for lookup_table_op
8 years ago
0bd710fcde
correct the init op type
8 years ago
chenweihang
6a05ed3366
fix: add batch size check in loading data process
8 years ago
fa94261b73
avoid duplicated init op for one parameter
8 years ago
690625fe15
Merge pull request #12456 from jacquesqiao/add-profiler-to-pserver
...
Add profiler to pserver
8 years ago
Jiabin Yang
Yancey1989
gongweibao
dzhwinter
Jacek Czaja
Michal Gallus
Qiao Longfei
chengduoZH
Bai Yifan
Xin Pan
Haihao Shen
qingqing01
minqiyang
Wu Yi
luotao1
tensor-tang
Yan Xu
Yu Yang
Qiyang Min
chengduo
tangwei12
Brian Liu
Chen Weihang
zenghsh3
whs
jerrywgz
fengjiayi
Xingyuan Bu
Yibing Liu
robot
Zeng Jinle
sneaxiy
dzhwinter
guochaorong
Michał Gallus
Qingsheng Li
yuyang18
Sylwester Fraczek