3be6c736f6
add multihead_attention.
8 years ago
d76fcb6fad
Memory optimization on Dynamic RNN ( #7599 )
...
* limit variable type to lod tensor in memory optimization transpiler
* refine policy
* support while operator
* fix random seed and training data order
* refine get_cfgs method to support multi while operators
* refine codes
8 years ago
c0ed2e4157
Change 0 to 0.0
8 years ago
f6a4c3ea77
Merge pull request #7779 from reyoung/feature/disable_detection_output
...
Disable the unit test ofdetection output op
8 years ago
81be9cef47
Merge pull request #6150 from wanghaox/prior_box
...
prior box operator for ssd
8 years ago
e845a1d5be
Enhance image.py for gray image.
8 years ago
25a0193215
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into iou_sim
8 years ago
c648244445
Merge pull request #7766 from guoshengCS/add-python-GRU
...
Add python wrapper for GRU
8 years ago
d498d93cf9
Fix comments and remove param_attr
8 years ago
1ee308623e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into block_expand_py
8 years ago
6f7b346f26
Merge branch 'develop' of github.com:baidu/Paddle into 7724
8 years ago
813d5b3c10
Disable the unit test ofdetection output op
8 years ago
ee21f2f601
Change default value of drop_rate in img_conv_group to 0
8 years ago
534cf74119
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into prior_box
8 years ago
a05d25cf41
update code and doc, change input x to LoDTensor
8 years ago
dace68ac6d
Merge branch 'develop' into multihead_attention
8 years ago
deb7509820
Merge branch 'develop' of github.com:baidu/Paddle into feature/wrap_nce_loss
8 years ago
05a733b025
Fix unit test bug in test_profiler.py.
8 years ago
59bcb5893e
Merge pull request #7759 from qingqing01/bipartite_match_op_fix
...
Fix bug and unit test in bipartite_match_op.
8 years ago
113cd6b31c
add multi-head scaled_dot_product attention.
8 years ago
69074eadbb
Merge branch 'develop' of github.com:baidu/Paddle into feature/wrap_nce_loss
8 years ago
8cfb3e55b4
Add python wrapper for GRU
8 years ago
575289209f
Fix bug and unit test in bipartite_match_op.
8 years ago
ce87b28970
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-7555
...
Conflicts:
python/paddle/v2/fluid/layers/nn.py
8 years ago
abf9395d45
Merge branch 'develop' into multihead_attention
8 years ago
eaabf2aa9f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into profiling_py
8 years ago
18a276c142
fix SyntaxWarning in wmt16.py
8 years ago
fdc59ae657
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into block_expand_py
8 years ago
d662e85f12
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into prior_box
8 years ago
d4587959f9
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into iou_sim
8 years ago
9b1a17a835
Refine conv2d_transpose layer doc ( #6920 )
...
* refine conv2d_transpose layer doc
* fix conv2d_transpose doc
* fix doc
8 years ago
0ed3f35919
pass parallel_do cbow
8 years ago
05e563b92e
Add unitest
8 years ago
cd25adbe27
Merge pull request #7659 from guoshengCS/enhance-reduceOp-datatype
...
Register reduce_op_kernel with multiple data types
8 years ago
2b19a68cc9
Merge pull request #7695 from qingqing01/bipartite_match_op
...
Add bipartite matching operator and unit testing.
8 years ago
479c861b57
Merge pull request #7726 from lcy-seso/fix_rendering_error_of_transpose_op
...
fix rendering error of transpose operator and add wrapper.
8 years ago
6d2cfe9220
Merge pull request #4866 from gongweibao/blockexpand
...
Add im2sequence op.
8 years ago
dcb5a1ed67
fix ci.
8 years ago
23f5c1829c
Fixed few comments in transpiler ( #7748 )
...
* Updating the cluster trainign doc
* Fixed few comments of transpiler
* Adding few explanations
8 years ago
09544bc1e6
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into blockexpand
8 years ago
a173fa7503
Merge pull request #7732 from JiayiFeng/refine_grad_clip_api
...
update gradient clip api
8 years ago
d51204425a
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into enhance-lookup_table_op-padidx
8 years ago
fbda30a2bc
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into my-cool-stuff
8 years ago
992f20ce7a
fix decode bug #7741
8 years ago
e44dedf90f
Fix the warning and unit test.
8 years ago
64606eafff
Merge branch 'develop' into fix_rendering_error_of_transpose_op
8 years ago
44561a24c9
Merge pull request #7655 from wanghaoshuang/ctc_evaluator_py
...
Add python wrapper for CTC greedy decoder and edit distance evaluator
8 years ago
5fc498e742
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_grad_clip_api
8 years ago
b3de0d9c17
Merge branch 'develop' into fix_rendering_error_of_transpose_op
8 years ago
9baba9a108
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into profiling_py
8 years ago
2edfd764ab
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-7555
8 years ago
74af23b6d1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into bipartite_match_op
8 years ago
22032e49cb
Add python wrapper for multiplex operator.
8 years ago
58cd4fda62
add wrapper for transpose operator.
8 years ago
d9d9be1bac
Fix white space in comments.
8 years ago
61d6db560b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-7555
8 years ago
38f47e642f
Fix CI
8 years ago
3e195d86d8
add wrapper for multihead_attention.
8 years ago
da0d95cb1c
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into blockexpand
8 years ago
8143a42667
1. Add more comments
8 years ago
b03c8b90ae
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into enhance-lookup_table_op-padidx
8 years ago
609ede25b9
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into profiling_py
8 years ago
579449b050
Update comments and revert pybind11.
8 years ago
1bc8de3209
1. Add sequence_num as edit distance op's output
...
2. Fix evaluator using 'reduce_sum' op instead of 'mean' op
8 years ago
89c591f37c
update grad clip api
8 years ago
336e8db1d1
Merge pull request #7729 from lcy-seso/fix_copyright
...
fix copyright
8 years ago
3c255ad39d
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into iou_sim
8 years ago
f7c0ad9d35
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into prior_box
8 years ago
528bcac52c
update iou_sim code
8 years ago
648ca7a3b2
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into blockexpand
8 years ago
c80af6ffaa
Merge pull request #7721 from reyoung/feature/rename_fluid
...
Move registery.py to layers module
8 years ago
54a4696f2a
Merge pull request #7660 from reyoung/feature/compare_op_use_elemwise
...
Make compare_op reuse elemwise_op_funcs
8 years ago
1c4968ee8b
fix copyright
8 years ago
af8cb820a7
Fix bug of nce_op
...
* also div num_samples when return cost of nce_op
8 years ago
142f632886
update code
8 years ago
3a48282e61
Fix unitest
8 years ago
41028f5746
Refine the doc.
8 years ago
5f1d2cb9cf
Fix typo.
8 years ago
430fdc52a8
Merge pull request #7661 from lcy-seso/wmt16_en_ger
...
Add WMT16 dataset.
8 years ago
a6da470b11
add memory optimization transpiler demo ( #7443 )
...
* add memory optimization transpiler demo
* add memory benchmark compile option
* add gflags instead of macro
* refine code
8 years ago
f020f4b51e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into prior_box
8 years ago
f035f327b1
Wrap NCE to python
...
And write an unittest for it
8 years ago
5c2318a989
Add unit test for python wrapper.
8 years ago
fedaa6563f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-7555
...
Conflicts:
doc/api/v2/fluid/layers.rst
python/paddle/v2/fluid/layers/nn.py
8 years ago
e8adcaf278
update
8 years ago
2b8ea2171e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_global_norm_clip
8 years ago
f45b0b0661
Merge pull request #7688 from reyoung/feature/python_overload_math_operators
...
Add math operator patches
8 years ago
5c26f60875
Fix license
8 years ago
d3d855fbc8
Fix unittest
8 years ago
3008aab6ee
Merge branch 'develop' of github.com:baidu/Paddle into feature/rename_fluid
8 years ago
1d715c6090
saml fix of lookup_table_op enhancement
8 years ago
2b9b6c3d32
Merge branch 'develop' of github.com:baidu/Paddle into feature/compare_op_use_elemwise
8 years ago
91b83d0a67
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into block_expand_py
8 years ago
87b424e866
Follow comments
8 years ago
4c7cb771aa
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into enhance-lookup_table_op-padidx
8 years ago
9d650112c0
Rename 'block_expand' to 'im2sequence'
8 years ago
ae676a609f
Enhance lookup_table_op to support padding_idx
8 years ago
2f344e7f0f
fix name convention.
8 years ago
18615626a8
Merge pull request #7715 from Yancey1989/split_selected_rows_to_multi_pserver
...
[WIP] Split SelectedRows to multiple pservers
8 years ago
0b854bdb8b
Add sequence_erase option into edit distance python API
8 years ago
f2c4bb679b
Add lstm with recurrent projection operator
8 years ago
eee62648cf
Refine test_label_semantic_roles of Fluid ( #7193 )
8 years ago
e983cc90fc
"fix decode bug" ( #7711 )
...
* "fix decode bug"
* "follow commnet"
* "fix error"
* "fix hook bug"
* fix based comment
* fix copyright
* fix based on comment
8 years ago
04806ffe83
Merge pull request #7656 from chengduoZH/feature/enhance_matmul_op
...
Enhance matmul_op to support 4-D inputs.
8 years ago
d0a9393668
Split SelectedRows to multiple pservers
8 years ago
782ddc5f59
follow comments
8 years ago
a8f118ca83
Add EditDistance to evaluator.py
8 years ago
680aec2156
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into ctc_evaluator_py
8 years ago
e92fe9e64f
Merge branch 'develop' into remote_optimize
8 years ago
cd38e2d142
remove conflict
8 years ago
ce93eea88f
refine unit test
8 years ago
2ac46d538d
Add distribution implement of image classification. ( #7687 )
...
Add distribution implement of image classification
8 years ago
19c554f9e4
update
8 years ago
c2edcde1b1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into bipartite_match_op
8 years ago
bf33b191d0
Add bipartite matching operator and unit testing.
8 years ago
811c4ee401
Add python wrapper for sequence_reshape.
8 years ago
4f933312db
Merge pull request #7662 from pkuyym/fix-6678
...
Add sequence reshape operator
8 years ago
9a97c7f745
add wmt16 into dataset.
8 years ago
c198523885
Merge branch 'fix-6678' of github.com:pkuyym/Paddle into fix-7691
8 years ago
22662ae424
Move paddle.v2.fluid.registery to layers
...
* registery is only used by layers
* Rename it to layer_function_generator
8 years ago
95b896ce72
update python interface
8 years ago
538f1ad28f
tiny fix
8 years ago
408a6b8bb2
tiny fix
8 years ago
42b0748ab4
add unittest
8 years ago
25dec82f24
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into ctc_evaluator_py
8 years ago
5846aab317
1. Rename 'edit_distance_error' to 'edit_distance'
...
2. Add edit distance evaluator to evaluator.py
8 years ago
55ec0e2a92
Merge pull request #7649 from wanghaoshuang/fix_edit
...
Change input data type to int64_t
8 years ago
9f731a608b
Add math operator patches
...
Users can use `a+b`, `a*10`.
8 years ago
47753a9667
Merge pull request #7527 from wanghaoshuang/ctc_greedy_decode
...
Add CTC align op
8 years ago
08cb472ab9
Simplify the implementation.
8 years ago
a1c281f07e
Merge pull request #7603 from kuke/simplify_erase
...
Enhance GPU kernel of sequence erase op
8 years ago
c5a14ed4cd
add mine_hard_examples operator
8 years ago
9c0b29014c
Make compare_op reuse elemwise_op_funcs
8 years ago
41b8388492
Merge pull request #7640 from kuke/add_lstm_doc
...
Add python doc for dynamic_lstm
8 years ago
8809d43ad7
Remove unnecessary dtype conversion & register int64 kernels
8 years ago
259fcc3026
Merge pull request #7676 from tonyyang-svail/7673
...
bring back test_parallel_do
8 years ago
9d91ae5f04
Merge pull request #7669 from putcn/develop
...
update doc and dist test due to transpiler change
8 years ago
58be41fa91
Merge pull request #7608 from typhoonzero/distributed_split_selectedrows
...
Enhance distributed train performance
8 years ago
07cf302e6e
first commit
8 years ago
974183b4da
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_global_norm_clip
8 years ago
b7b5de7fd6
Merge pull request #7665 from JiayiFeng/dev_update_auto-registry
...
make auto-registry layers supporting specified output
8 years ago
c79d530ad3
Add split selected rows op ( #7604 )
...
* add split selected rows op
* update comment
* add grad check
* registry cuda kernel
* fix ci failed
8 years ago
0aff136395
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into distributed_split_selectedrows
8 years ago
95d6dce354
update doc and dist test due to API change https://github.com/PaddlePaddle/Paddle/pull/7619#pullrequestreview-89656079
8 years ago
7a68787667
Merge pull request #7667 from abhinavarora/dist_transpiler
...
Remove optimize_op argument from get_pserver_program method of Distributed Transpiler
8 years ago
030c0b91fb
Do not send to optimize_ops to distribute transpiler again
8 years ago
d77e6a6760
Merge pull request #7636 from kexinzhao/save_inference_model
...
Add feed and fetch op to ProgramDesc before saving for inference
8 years ago
7905e36741
Implement Book chapter 02 in distributed framework ( #7629 )
8 years ago
84de7e7f03
make auto-registry layers supporting specified output
8 years ago
773f2f735c
fix errors
8 years ago
ef56e6839a
Correct the usage of fc in the example of dynamic_lstm's doc
8 years ago
01d568e59a
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into ctc_evaluator_py
8 years ago
4673a4a9aa
divide this operator into ctc_greedy_decoder and edit_distance_error.
8 years ago
bea41444d7
Refine the implementation and add unit test.
8 years ago
30529e314e
delete debug transpiler code
8 years ago
6e04e580d2
Change input data type to int64_t in unitest and GPU kernel
8 years ago
b83ff4514b
Register reduce_op_kernel with multiple data types
8 years ago
2edc136c5f
add 4-d for matmul_op
8 years ago
2360dd20a4
Merge pull request #7438 from wanghaoshuang/ctc_py
...
Add python API for Warp-CTC op
8 years ago
2ca603b98f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into ctc_evaluator_py
8 years ago
082c302c3f
Add comments
8 years ago
3b0eff6196
Format the writing in doc of dynamic_lstm
8 years ago
f46257fa4a
Merge branch 'develop' of upstream into add_lstm_doc
8 years ago
f050390754
Polish the doc of dynamic_lstm
8 years ago
0dd3919a21
Add python wrapper for ctc_evaluator
8 years ago
51985aa2aa
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_global_norm_clip
8 years ago
4de6cbd308
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into ctc_py
8 years ago
edd2132679
remove conflict
8 years ago
856f650a9b
fix bug
8 years ago
4b3e22b865
Merge pull request #7574 from lcy-seso/wraper_for_l2_normalize
...
add python wrapper for l2 normalize layer.
8 years ago
1d89866995
Merge pull request #7593 from JiayiFeng/dev_elementwise_scalar
...
Make elementwise_op supporting scalar input `Y`
8 years ago
37a9437073
Merge pull request #7538 from JiayiFeng/dev_elementwise_max_min
...
elementwise max min
8 years ago
6edbf1387c
remove ptools
8 years ago
e1f475ad1e
refine code
8 years ago
e469545757
Add Copyright to test_ctc_align.py
8 years ago
ed3e5717ba
fix bug
8 years ago
dc168ed00a
modify programDesc based on feed and fetch names
8 years ago
151f579c23
Recv OP: use BlockDesc* instread of ProgramDesc proto as Attribute
8 years ago
b8a17987ec
Feature/parallel for bug fix ( #7474 )
...
* Fix ParallelDo not support empty input gradient
* Polish ParallelDo and fix several bugs
* Fix CI
* Fix CI
8 years ago
193096194c
Merge branch 'dev_elementwise_max_min' into dev_elementwise_scalar
8 years ago
a37f6ad379
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_elementwise_max_min
8 years ago
e043c2ce45
Merge branch 'develop' into wraper_for_l2_normalize
8 years ago
a247972dda
fix a error
8 years ago
958d07bee3
fix a error
8 years ago
1dac173b51
add API for clip_by_global_norm
8 years ago
f233b936c7
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into distributed_split_selectedrows
8 years ago
6ebfade465
fix copyright information
8 years ago
3f5c77cc4b
fix copyright information
8 years ago
fe45f2115f
1. Rename 'block_expand' to im2sequence
...
2. Refine code and doc
8 years ago
f1a889720a
fix copyright
8 years ago
4cb6e72b85
refine code details
8 years ago
02ea349101
enhance dist train performance
8 years ago
5ae0c97faf
Add unit test case for no tokens to be erased
8 years ago
db959d6350
Add dot_product_attention to nets.__all__
8 years ago
66054984cd
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-dot_product_attention
8 years ago
228e14adb7
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_global_norm_clip
8 years ago
f189ad7442
refine the defination of class GradientClipByGlobalNorm
8 years ago
9bcb2d268e
Add python wrapper for matmul_op and dot_product_attention
8 years ago
7150289b5c
Refine CPU kernel
...
1. Allocate memory for output before compute.
2. Rename 'ctc_decode' to 'ctc_align'
8 years ago
14f6fa346b
make elementwise op support scalar as input Y
8 years ago
11b4471ade
Add python API for block expand op
8 years ago
ed7e74abf2
follow comments and refine python doc
8 years ago
87a59d65d6
Merge branch 'develop' into wraper_for_l2_normalize
8 years ago
24f528a1a5
follow comments
8 years ago
adcfde3eab
Modify unitest
8 years ago
3177c9a691
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_elementwise_max_min
8 years ago
bfe7e24243
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into blockexpand
8 years ago
d15bfabbd0
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_elementwise_max_min
8 years ago
e82f1008a8
Finish block expand op
...
1. Add lod to output
2. Fix im2col arguments list
3. Refine code and doc
4. Fix output shape
8 years ago
aab4cfeb65
Add doc for dynamic_lstm python api
8 years ago
d2a70243f1
Refine profiler and expose to Python.
8 years ago
2a0a576130
Add non-linear activation.
8 years ago
8266fcc3be
Add pyton wrapper for row conv operator.
8 years ago
3b63815629
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into iou_sim
8 years ago
2ad5a6f0d1
add iou similarity operator
8 years ago
6497bff901
add python wrapper for l2 normalize.
8 years ago
10dd632659
Rename 'ctc_greedy_decode' to 'ctc_decode'
8 years ago
281e93bcbb
Remove 'top 1' from CPU and GPU kernel
...
1. Remove 'top 1'(or argmax) from CPU and GPU kernel
2. Add a new test case
3. Refine doc
8 years ago
4031e88dc4
"fix the copyright hook"
8 years ago
1511a04934
Merge pull request #7540 from pkuyym/fix-7533
...
Add reorder flag for DynamicRNN's memory function.
8 years ago
d968b4ba1a
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into ctc_py
8 years ago
24144bdb2c
Add notice about norm_by_times and change 'ctc' to 'CTC'
8 years ago
f23691db4b
Merge pull request #7434 from pkuyym/fix-7195
...
Add static_input for DynamicRNN
8 years ago
448fee3db4
Merge pull request #7414 from wanghaoshuang/warpctc
...
Adapt warpctc grad op for gradient checking
8 years ago
c01bb26f1d
Add reorder flag for DynamicRNN's memory function.
8 years ago
b9b75377a2
Feature/hooks ( #7513 )
...
* add copyright hook
* add copyright hook
* refine copyright hook
* "test copyright hook"
* fix check style
* fix ci
8 years ago
251c6032fb
set use_cudnn as default
8 years ago
bcfb82d33e
dist train support split selectedrows
8 years ago
ee8e5374d8
add max min layer
8 years ago
c6d7ad3689
fix typo
8 years ago
1e6e5ac64c
add unit test
8 years ago
78dc93430c
expose use_cudnn
8 years ago
f2e1f3e114
Fix activation for glu
8 years ago
234013a9d7
Add python wrapper for matmul_op
8 years ago
e7acf32cdf
Merge pull request #7525 from guoshengCS/add-python-glu
...
Add python wrapper for GLU.
8 years ago
8f37c3c2a7
Fix sequence scale functor cuda kernel
...
1. Fix kernel
2. Add more test case
8 years ago
25fee87154
Change program.seed from 0 to 1.
8 years ago
0237b7e99a
"remove random shuffle" ( #7521 )
8 years ago
3bbff25bc3
Merge pull request #7453 from reyoung/feature/fix_seed_for_dynrnn_test
...
Fix random seed of dynamic rnn gradient check
8 years ago
579f684661
Add ctc_greedy_decode_op
8 years ago
f05330b78b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-7195
8 years ago
ef129718ea
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-python-glu
8 years ago
8d253e4965
Merge pull request #7249 from typhoonzero/transpiler_split_tensor
...
Feature/transpiler split tensor to multiple pservers
8 years ago
c083a60d7a
Add python split and glu
8 years ago
e5c89143c4
Merge pull request #7431 from JiayiFeng/dev_error_clip_test
...
ErrorClip test on MNIST
8 years ago
a091d1a31c
Enhance print_op.
8 years ago
c996eb8a0b
rename dist tests
8 years ago
0a44cd91df
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_error_clip_test
8 years ago
adc26dffa9
developing GradientClipByGlobalNorm
8 years ago
59bc4c4600
fix dynamic rnn bug in GPU ( #7480 )
8 years ago
b886b8292d
add cmakelist
8 years ago
5ad1aef051
"cudnn operators change to cudnn kernel" ( #6660 )
...
* "unified operators"
* "add CUDNN register"
* "add use cudnn attribute"
* "add attribute"
* "test conv tranpose op"
* "remove duplicated attr"
* "fix op test"
* "add attribute to set cudnn"
* "add more log"
* "need layout op register support"
* "add more log"
* "change GetExpectedKernelType "
* "fix Get attr in conv_op"
* "fix CI"
* "fix tests"
* "removed kernel priority fallback"
* "fix CI"
* "fix stack pointer bug"
* "refine buggy interface"
* "add const cast to save life"
* "fix get_output_with_grad"
* "fix op test with dataformat"
* ""fix pooling
* "fix pooling test"
* "fix CI"
* "fix with_gpu error"
* "add transform needed functional check"
* "fix unpack list error"
* "comment out parallel.do temporary"
* "fix CI"
* "fix compile doc error"
* "make threshold larger"
8 years ago
45cf234127
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into warpctc
8 years ago
137f0dfc21
1. Fix warpctc grad tensor initial bug.
...
2. Remove num_seq arguments.
3. Refine CUDA kernel of ScaleLoDTensorFunctor.
4. Change max_relative_error of gradient unitest to 0.007
8 years ago
df9c13a816
Merge pull request #7485 from lcy-seso/wrapper_for_elementwise_math_op
...
add wrapper for element-wise math operator.
8 years ago
8ab611d559
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into prior_box
8 years ago
1ba3d29789
update code
8 years ago
db65f497ff
Update comments for two operators. ( #7457 )
...
* update code comments.
* update the comments.
* follow comments.
8 years ago
8ac744f372
add wrapper for elementwise math operator.
8 years ago
df50c14abd
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into transpiler_split_tensor
8 years ago
06b326b681
follow comments
8 years ago
d23ea4ef8e
add gradient clip by norm
8 years ago
c24da0d3ee
update unit test split var
8 years ago
697f92ef86
Add type check
8 years ago
3423022e84
feature/add print op ( #6799 )
8 years ago
b4d1811c9a
Merge pull request #7470 from kavyasrinet/doc
...
Adding the rst file to render IO under Fluid in paddlepaddle.org
8 years ago
5faebab375
Done, need support selectedrows
8 years ago
d954becb1c
Merge pull request #7371 from emailweixu/assign_value_op
...
assign_value operator
8 years ago
54791af658
Addressing review comments
8 years ago
c806eeff61
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into doc
8 years ago
25c8739362
Adding the rst file to render IO under Fluid in paddlepaddle.org
8 years ago
cadb95fbeb
minor tweaks
8 years ago
b1f24660bf
add dist demo fit_a_line
8 years ago
5d9dcfc1d0
Merge pull request #7429 from putcn/book_demo_distributed_understand_sentiment_
...
Book demo understand sentiment distributed version
8 years ago
bfc68f256a
Merge pull request #7425 from helinwang/distributed_label_semantic_role
...
Add distributed label semantic role book chapter
8 years ago
237385cf41
Correctly handle int values for assign_value_op
8 years ago
ce233796ea
assign_value operator
...
We need this operator to assign value to a tensor and the values are stored in the program so that they can be used independent of python.
8 years ago
6c0723661e
Add distributed label semantic role book chapter
8 years ago
ea782e38a6
Fix typo in batch norm bias initialization ( #7449 )
8 years ago
5d901d00bf
update
8 years ago
5325313e4c
debugging shape match
8 years ago
a9dbdab5e2
Merge pull request #7396 from reyoung/feature/parallel_for_unittest
...
Feature/parallel for unittest
8 years ago
e5b23785fe
Fix random seed of dynamic rnn gradient check
8 years ago
95c0c12641
Merge pull request #7384 from dzhwinter/feature/sync_wait
...
Feature/sync wait
8 years ago
a7c2bfb4a1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-7195
8 years ago
83c72536e6
Update batch_size
8 years ago
fd24e195a3
Uncomment check output in unitest
8 years ago
2fd7675cd3
Add lod checking for forward.
8 years ago
fb62f8cb0f
Add python api for warp-ctc op
8 years ago
1797f3db85
Refine memory optimization transpiler ( #7394 )
...
* add update graph method for memory optimization transpiler to avoid rebuild graph everytime
* clean code
* reset var desc if hit cache
8 years ago
92eb247f07
"fix stupid error"
8 years ago
6f79bbc93e
Add unittest except gradient checking (since bugs).
8 years ago
dc488c17d1
Merge branch 'develop' of github.com:baidu/Paddle into feature/parallel_for_unittest
8 years ago
12aca860bf
Add comment
8 years ago
87f9b58363
set stop gradient for mask in dropout layer ( #7390 )
8 years ago
0064f6d976
refine test
8 years ago
59bf85d914
Merge pull request #7325 from kuke/sequence_erase_op
...
Add sequence erase op
8 years ago
9eb3fb2946
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into warpctc
8 years ago
b1af5e435f
1. Fix warpctc grad op
...
2. Add check grad test
8 years ago
2827607fa8
fix startup program shape
8 years ago
af96cd5877
Merge pull request #7397 from luotao1/openblas_env
...
auto set openblas env
8 years ago
6fa56b9d01
left startup program bug
8 years ago
52dad013ce
Add static_input.
8 years ago
f3dcd00603
Merge branch 'develop' of upstream into ctc_edit_distance_dev
8 years ago
6cff3c96db
Merge pull request #7381 from reyoung/feature/refine_get_places_op
...
Polish GetPlacesOp
8 years ago
1f5f79cbf5
fix error_clip errors and add unit test
8 years ago
a1935b23c4
Remove unnecessary prefix in test name of edit_distance_op
8 years ago
a32027604f
Merge pull request #7176 from pkuyym/fix-7173
...
Enhence shrink_rnn_memory_op.
8 years ago
929d22c622
auto set openblas env
8 years ago
2412f2f412
Polish Unittest for ParallelFor
8 years ago
c7eb199b2e
Init commit
8 years ago
10779460c5
Simplify calc in test_sequence_erase_op
8 years ago
e5e206e2b6
Merge branch 'develop' of github.com:baidu/Paddle into feature/refine_get_places_op
8 years ago
4bcc0b64cb
[WIP] feature/parallel_gpu ( #7293 )
...
feature/parallel_gpu
8 years ago
42daf4c302
"add sync op"
8 years ago
4059c9ca7f
Polish GetPlacesOp
8 years ago
df92776850
Merge pull request #7269 from emailweixu/calc_gradient
...
Calculating gradients for partial graph
8 years ago
5f98500009
Make init device on all gpu by default ( #7345 )
...
* "init use all default devices"
* "fix init test"
8 years ago
efe06caa3d
change data type of beam_search op ( #7374 )
8 years ago
50a02adf5e
transpile program ok
8 years ago
6e5eae137d
Unittest for fluid.backward.calc_gradient()
8 years ago
585dec3dc2
Calculating gradients for partial graph
...
Added backward.calc_gradient to backpropagate gradient from given targets to inputs.
8 years ago
4a3580ac75
add missing dependency
8 years ago
6ecbf08372
add memory optimization transpiler ( #7356 )
8 years ago
9c0b1cf1c2
update wip pserver transpile
8 years ago
99a6c5d40e
change output shape to [2, layer_height, layer_width, num_priors, 4]
8 years ago
e249ad1211
Test dist word2vec ( #7334 )
...
* test dist word2vec
* multiple trainers work
8 years ago
b5fda2723f
Port WarpCTC Operator ( #5107 )
...
* Add Seq2BatchFunctor, which will be used in WarpCTCOp.
* Implement WrapCTCFunctor and WrapCTCKernel.
* Add unittest of warpctc_op.
* Modify the check_output inferface in python unittest framework to allow check a subset of outputs.
* Use absolute offset lod in warpctc_op and related functors.
* Refine the comments of warpctc_op.
* The new python unittest supports checking a subset of the outputs, so revoke the previous change.
* Rename the transform from LoDTensor to Tensor with shape [max_sequence_length, num_sequences, sequence_width] to PaddingSequenceFunctor.
* Update to the newest codes.
* Rename the PaddingSequenceFunctor to PaddingLoDTensorFunctor and remove the computation of dimensions out of the functos.
8 years ago
fe341bacde
refine batch norm python layer ( #7348 )
8 years ago
56e758fc10
trainer ok
8 years ago
6ece41ecb9
Merge pull request #7213 from JiayiFeng/dev_add_callback_for_backward
...
Error Clip Design Doc
8 years ago
452fe1e1be
init checkin for distributed book chapter 6
8 years ago
7b9d5b325c
Add document for sequence_erase_op
8 years ago
37f933b8ad
Add gpu kernel for sequence_erase_op
8 years ago
f35c56060c
split tensor to pservers
8 years ago
12ed53c1f2
Inherit LoD from x to x_grad and enhance the unit test.
8 years ago
41307c738d
Merge pull request #7284 from pkuyym/fix-7211
...
Stop gradient when pool_type=='max'
8 years ago
219fbd51bc
Merge pull request #6732 from QiJune/get_places_op
...
add GetPlaces operator
8 years ago
2a54ddd267
Merge branch 'develop' of upstream into sequence_erase_op
8 years ago
26d451116b
Merge pull request #7251 from guoshengCS/enhance-reorderTensor
...
Enhance reorder_lod_tensor_by_rank_op to support Tensor
8 years ago
67b8c09210
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-7211
8 years ago
63ff0b4ba9
Refine get_places
8 years ago
872b1c8884
Stop gradient when pool_type=='max'
8 years ago
0f353ab46e
cpu gpu transform function ( #7191 )
...
* add rename guard
* add device_data_transform
* add device_data_transform_test
* modify GetExpectedKernelType
* update operator.run
* support test test_label_semantic_roles
* optimize code
* optimize code
* rename GetActualKernelType to GetExpectedKernelType
* fix chunk_eval_op and device_data_transform_test
* add is_same_place to place
* optimize code, refine rename_guard
* refine rename guard, add GetKernelTypeForVar
* optimize code
* add some log
* rename guard
* use sub scope to create var
* fix compile
* add IsInitialized for Tensor
* add VarIsTensor
* fix op_registry_test
* test
* tmp disable priority
* restore switch_kernel.md
* code clean
8 years ago
a9f7cd340b
Merge pull request #7278 from pkuyym/fix-7216
...
Bug fix when inserting fill_zeros_like_op.
8 years ago
ed0cf3d6c6
Merge branch 'develop' of github.com:baidu/Paddle into feature/get_places
8 years ago
691b5cac61
Fix equation for gru op ( #7274 )
8 years ago
758fe4731c
Merge pull request #7260 from emailweixu/conditional
...
Fix the output of conditional_block
8 years ago
05c08214e3
Bug fix when inserting fill_zeros_like_op.
8 years ago
12e3514139
Modify inference.cc to run example without pickletools ( #7262 )
8 years ago
e85c513307
Add sequencee erase operator
8 years ago
0b6c5e6d25
Merge pull request #7223 from PaddlePaddle/emailweixu-patch-1
...
Fix comment to layers.fc
8 years ago
787d1d36d0
Fix the output of conditional_block
8 years ago
ea6eb96347
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into enhance-reorderTensor
8 years ago
e21923543a
Enhance reorder_lod_tensor_by_rank_op to support Tensor
8 years ago
c70ea1cc30
add splitter
8 years ago
ed55f1b9d4
transpiler_split_tensor
8 years ago
be218bfac4
fix bug
8 years ago
33cb12a89b
update error clip
8 years ago
dea5263179
update error clip
8 years ago
894236a128
Merge pull request #6730 from tonyyang-svail/parallel_do
...
[WIP]: feature/parallel_do
8 years ago
5593858dd9
Feature/use cudnn ( #7141 )
...
* "add c++ side kernel selection"
* "add multiple kernel op test"
* "kernel selection only support cudnn"
* "better formatter"
* "small fix with UseCPU"
* "depends on change interface Get(Place, Library)"
* "fix CI"
* "fix python cudnn test"
* "leave the register cudnn op to another PR"
* "fix CI"
* "use all kernel by default"
* "fix CI"
8 years ago
8496b2e41b
Refine parallel_do
8 years ago
60e27d110b
Merge branch 'develop' of github.com:baidu/Paddle into parallel_do
8 years ago
809122cb5a
Merge pull request #7097 from Xreki/core_inference_example
...
Add a simple C++ inference example for fluid
8 years ago
f3c42f607c
Add doc for gru_unit op (in fluid) ( #7151 )
...
* Add squared error layers doc
* Add doc for gru_unit
* Remove cdot which isn't supported
* Update layers.rst
* Update layers.rst (minor)
8 years ago
ebc616b0f9
Fix comment to layers.fc
...
x_num_col_dims => num_flatten_dims
8 years ago
2b259bf14d
Merge pull request #7208 from reyoung/feature/default_omp_num_threads_1
...
Default use one thread in fluid
8 years ago
4ead8e1b57
Add doc for error clip
8 years ago
ee341efd79
Merge pull request #7183 from tensor-tang/use_mkl_packed
...
Add flag use_mkl_packed
8 years ago
a4024a5f3d
"remove cudnn devicecontext" ( #7207 )
...
* "remove cudnndevicecontext"
* "remove unused init code"
* "fix hash functions"
8 years ago
b585c93ab0
Default use one thread in fluid
8 years ago
134c5c4db7
Support callback
8 years ago
cd5fad13ce
Merge pull request #7160 from reyoung/feature/expose_activations
...
Expose some activations
8 years ago
f947c15378
Consider multiple levels of LoD.
8 years ago
042f3524d2
add flag use_mkl_packed
8 years ago
89bbc4f6e5
Merge pull request #7157 from pkuyym/fix-7156
...
Doc fix and enhancement for lstm_unit python wrapper.
8 years ago
66ae0a8cb2
Enhence shrink_rnn_memory_op.
8 years ago
7411df34b8
add multi thread
8 years ago
8ee17e965f
pass sgd at first iter
8 years ago
9313233297
merge develop
8 years ago
0250e54c2d
Enable batch input in edit_distance_op
8 years ago
0c16f4f602
Update
8 years ago
cbe25b3345
Merge pull request #6881 from sweetsky0901/cross_channel_norm
...
add norm op（cross_channel_norm）
8 years ago
059096741e
Add init glog
8 years ago
60fecce43d
Fix unit test for lstm_unit.
8 years ago
90a5a55a6c
Expose some activations
8 years ago
42a0603e6f
Merge branch 'develop' into core_inference_example
8 years ago
c0f6f492bc
Add shape info for arguments.
8 years ago
e57a40b8a6
Merge pull request #7140 from JiayiFeng/dev_refine_backward_no_grad_var_handling
...
Dev refine backward no grad var handling
8 years ago
d6ec963047
Minor correction.
8 years ago
f0e797e5b7
Doc fix and enhancement for lstm_unit python wrapper.
8 years ago
87f46ebb36
Add squared error layers doc ( #6862 )
8 years ago
27fea24fd1
Addign document for fluid split_lod_tensor and merge_lod_tensor ( #6859 )
...
* Addign document for fluid split_lod_tensor
* Adding document for fluid merge_lod_tensor
8 years ago
e9a60e4c8e
Adding API docs for ones and zeros methods ( #7150 )
8 years ago
f3812825d0
Added documentation for topk ( #6861 )
8 years ago
06888bb05f
Merge pull request #7132 from pkuyym/fix-7023
...
Add doc for max_sequence_len.
8 years ago
33e75201e9
fix bugs
8 years ago
1bcf7e23bd
Minor refinement.
8 years ago
8d4a607fb3
update backward doc
8 years ago
7be57de943
enhance no_grad_var handling
8 years ago
9cfa5ce30f
Merge pull request #7128 from pkuyym/fix-7024
...
Add doc for lod_rank_table.
8 years ago
0219c3d6d2
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cross_channel_norm
8 years ago
0d4fdce07f
Minor refinement.
8 years ago
cd7d0f850b
Merge branch 'develop' into core_inference_example
8 years ago
a5200b89ac
Add doc for max_sequence_len.
8 years ago
57bc564d12
Polish doc for lod_rank_table.
8 years ago
10cd6eb67a
Add doc for lod_rank_table.
8 years ago
59c14f0b6e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into detection_output
8 years ago
6a5cf28a7f
Merge pull request #7123 from JiayiFeng/Add_backward_doc_and_comments
...
Add backward doc and comments
8 years ago
8b91174c83
Merge pull request #7027 from reyoung/feature/rnn_gradient_check
...
Feature/rnn gradient check
8 years ago
0bd7f97b4b
Merge pull request #7045 from typhoonzero/adam_selectedrows
...
Adam selectedrows and scatter functors
8 years ago
2e49facae9
Rename inputs & format license
8 years ago
afb5496c16
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into detection_output
8 years ago
8543ad6463
update backward doc
8 years ago
2c1adb0604
Rename ctc_edit_distance_op to edit_distance_op
8 years ago
fcd84c1530
Comment debug code
8 years ago
dd0a4c3566
Merge branch 'feature/check_nan_executor' into feature/rnn_gradient_check
8 years ago
d0cc7cc2b6
Merge branch 'develop' of github.com:baidu/Paddle into feature/check_nan_executor
8 years ago
641b4c0fe6
wip
8 years ago
9b3f2c39f2
Add a simple example for fluid to do inference in C++ code.
8 years ago
bb0427add0
Add comments for functions in backward.py
8 years ago
d2cb28413e
Fix ALL RNN error
8 years ago
dd2bbf3a14
update md5 of flowers dataset
8 years ago
96bc335216
Update
8 years ago
c732279359
Merge branch 'feature/check_nan_executor' into feature/rnn_gradient_check
8 years ago
dc7ddcb0b7
resolved conflict
8 years ago
5f3dd266f3
Merge branch 'feature/is_nan' into feature/check_nan_executor
8 years ago
3b54948668
Merge pull request #7073 from JiayiFeng/enhance_DataFeeder
...
enhance DataFeeder
8 years ago
3a93fa7708
Merge pull request #7069 from kuke/fix_cross_entropy_doc
...
Polish the doc of cross_entropy
8 years ago
f9a1229666
Merge pull request #6850 from chengduoZH/feature/conv2d_python_doc
...
Add conv2d_python doc
8 years ago
3d2b2d408f
refine doc
8 years ago
36ec3e9042
Merge branch 'develop' of upstream into ctc_edit_distance_dev
8 years ago
3db7c82955
Merge branch 'develop' of upstream into fix_cross_entropy_doc
8 years ago
b775b6cbaa
Merge pull request #6741 from JiayiFeng/dev_new_backward
...
[WIP] new backward
8 years ago
2f76932d7c
enhance DataFeeder
8 years ago
e2c2652fc0
amend comments in cross_entropy_op
8 years ago
6f5e64af17
Merge branch 'feature/check_nan_executor' into feature/rnn_gradient_check
8 years ago
5162c41a92
Add gflags
8 years ago
4745a0b9d8
Merge branch 'develop' of upstream into ctc_edit_distance_dev
8 years ago
4177e80545
Add line feed character in the doc of cross_entropy
8 years ago
c67c54a8e7
Polish the doc of cross_entropy
8 years ago
1cb9635947
fix dist train trainspiler bugs
8 years ago
dd21ae6c1e
update
8 years ago
14b87dbfeb
update reorder_lod_tensor_op_test
8 years ago
c532fdab29
fix errors
8 years ago
5361911c68
adam support sparse
8 years ago
bff0cbfcd3
Merge pull request #7025 from JiayiFeng/rename_output_of_softmax_and_activitions
...
Rename output of softmax and activations
8 years ago
073746f549
Merge pull request #7036 from JiayiFeng/dev_add_type_check
...
Add data type checking for io.save_inference_model()
8 years ago
5b9dbbb983
code clean
8 years ago
ying
QI JUN
wanghaoshuang
Yu Yang
Wang Hao
qingqing01
Guo Sheng
guosheng
yangyaming
Luo Tao
chengduo
Yang Yang
kavyasrinet
fengjiayi
yeyupiaoling
武毅
Yibing Liu
ranqiu92
dzhwinter
Yancey1989
chengduoZH
helinwang
gongweibao
Yancey
Xi Chen
Abhinav Arora
kexinzhao
Abhinav Arora
Kexin Zhao
caoying03
dzhwinter
Yan Chunwei
emailweixu
xuwei06
yi.wu
Qiao Longfei
Yiqun Liu
Siddharth Goyal
tensor-tang
sweetsky0901
sweetsky0901