Qiao Longfei
be9a2b2e30
Merge pull request #8652 from jacquesqiao/fix-parallel-do-with-non-parameter
...
parallel-do should not merge the gradient of parameter that stop_gradient=True
7 years ago
Yu Yang
22b5c07a7d
Fix the compilation on CUDA 9.1/GCC 5.3
...
* Make CUPTI_LIB_PATH not passing by macro.
* Add missing header
7 years ago
wanghaoshuang
8d57e9c76f
Rename 'edit distance rate' to 'edit distance'
7 years ago
wanghaoshuang
795f572fee
Rename 'seq_error' to 'instance_error'
7 years ago
fengjiayi
101378c878
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into remove_evaluator
7 years ago
qiaolongfei
c4996fb801
Merge branch 'feature/parallel_nccl_not_support_batchnorm' of https://github.com/reyoung/Paddle into fix-parallel-do-with-non-parameter
7 years ago
qiaolongfei
f8d0d84f7e
fix multi card
7 years ago
qingqing01
f9c1b6f9ae
Fix the backward transpiler bug in ssd_loss API. ( #8648 )
7 years ago
Yibing Liu
ace512a3ed
Merge pull request #8596 from kuke/profiler_multi_gpu
...
Fix the profiler's bug in multi-gpu mode
7 years ago
Tao Luo
62698d6e17
Merge pull request #8634 from typhoonzero/fix_renamevar_ref
...
Fix rename_var de-ref
7 years ago
qiaolongfei
5d305070ec
parallel-do should not merge the gradient of parameter witch set stop_gradient=True
7 years ago
Yu Yang
77ff14aa78
Not support batchnorm
7 years ago
wanghaoshuang
1e510d9914
Add ceil_mode option for pool2d and pool3d
7 years ago
Yibing Liu
c0876cf686
update due to upstream's change
7 years ago
chengduo
81374eb15e
Merge pull request #8644 from chengduoZH/doc/FAQ_refine
...
refine FQA doc
7 years ago
chengduoZH
6019aba67a
refine FQA doc
7 years ago
fengjiayi
d01318c0be
add WeightedAverage
7 years ago
dzhwinter
313454dfae
"init"
7 years ago
wanghaoshuang
87d90d2afd
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into seq_error
7 years ago
typhoonzero
1e9efc2272
add comments
7 years ago
chengduoZH
90dc33b5ff
Add todo for reduceSum
7 years ago
qingqing01
69643b5e89
Enable the SSD loss to support normalization by the total number of output locations. ( #8630 )
...
* Register more data type for reshape operator.
* Enable the SSD loss to support normalization by the total number of output locations.
* Fix the doc format.
7 years ago
typhoonzero
bd87882651
fix rename_var deref
7 years ago
Yu Yang
2a4e7add6f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/add_global_step
7 years ago
Yiqun Liu
f0548679b3
Get rid of the dependency of Go compiler when WITH_GOLANG is OFF. ( #8610 )
...
* Get rid of the dependency of Go compiler when WITH_GOLANG is OFF.
* Minor modification to the documentation.
* Remove the downloading and installing of Go compiler when build developing Docker image for android.
* Minor modifcation to the documentation's format.
* Change the default value of WITH_GOLANG to OFF in paddle/scripts/docker/build.sh.
7 years ago
Tao Luo
00595f7591
Merge pull request #8622 from gongweibao/fixdoc
...
Add cluster-train document.
7 years ago
gongweibao
6f4c2f0a46
follow comments
7 years ago
gongweibao
4ef4bd8e8e
clean up
7 years ago
chengduo
cf883d9caf
Merge pull request #8613 from chengduoZH/feature/fix_bug_conv_op
...
Fix conv_op bug
7 years ago
qingqing01
f449180b1c
Register more data type for reshape operator. ( #8617 )
7 years ago
gongweibao
97a132d5a9
init
7 years ago
chengduoZH
a779b42450
follow comments
7 years ago
xuwei06
e9b8ebf42c
Correctly handling variable with batch dimension for math ops.
...
When the second argument contains batch dimension, the axis should be 0.
Also makes elementwise ops more tolerant at handling tensors with trailing
singular dimensions.
7 years ago
xuwei06
7d56c6d078
Adding Dim<0>
...
Dim<0> is for scalar (rank-0 tensor). Adding Dim<0> can simplify
a lot of code.
7 years ago
Tao Luo
a67cebaf5f
Merge pull request #8615 from luotao1/conf.py.in
...
remove `import paddle.fluid` in conf.py.in
7 years ago
Luo Tao
0b73cb3358
remove `import paddle.fluid` in conf.py.in
7 years ago
chengduoZH
7cf2c05fd2
add unit test for input's size is 1x1
7 years ago
chengduoZH
b5c920920c
fix conv_op bug
7 years ago
Tao Luo
d742f4e54b
Merge pull request #8604 from luotao1/batch_size_like
...
combine batch_size_like.cc into batch_size_like.h
7 years ago
武毅
4d4322a698
merge fluid dist tests ( #8573 )
...
* merge fluid dist tests
* update cmake
7 years ago
Luo Tao
6dd3a61b2f
combine batch_size_like.cc into batch_size_like.h
7 years ago
Wang Hao
ec33832602
Merge pull request #8577 from wanghaox/python_map
...
add warrper for detection map operator
7 years ago
guru4elephant
daa8ed338f
Merge pull request #8495 from PaddlePaddle/fix_conll05_bug
...
Update conll05.py
7 years ago
wanghaox
52c5285412
Merge branch 'python_map' of https://github.com/wanghaox/Paddle into python_map
7 years ago
Yancey
c02f773a53
Fix dist demo var type error ( #8600 )
...
* Fix dist demo error
* revert trainer_id
7 years ago
Xin Pan
decaad5ccc
Merge pull request #8542 from panyx0718/test
...
Extend current profiler for timeline and more features.
7 years ago
gongweibao
ef35c4ed1a
Tensorflow benchmark ( #8522 )
...
Tensorflow benchmark
7 years ago
wanghaox
1e1202b659
merge detection.py
7 years ago
Wang Hao
672f20835f
Merge branch 'develop' into python_map
7 years ago
Yu Yang
5f70629ffa
Better function name
7 years ago