f7b768d364
fix group_norm ( #15727 )
...
* fix group_norm
* test=develop
6 years ago
25ec6a2cfd
Merge pull request #15657 from qingqing01/fix_bn_api
...
Fix batch_norm API for data_layout.
6 years ago
7e399b0628
rename
...
test=develop
6 years ago
4a1c65ebee
merge origin
...
test=develop
6 years ago
f85245b409
test=develop
6 years ago
01d9bf9264
Fix batch_norm API for data_layout.
...
test=develop
6 years ago
6f0f8045f6
Revert "Async double buffered py reader"
6 years ago
ceb412b0ae
speed up box coder in CPU, test=develop
6 years ago
db8fcf6b95
Merge pull request #15296 from cjld/async_double_buffered_py_reader
...
Async double buffered py reader
6 years ago
4eb44380a6
Merge branch 'develop' into add_clip_op
6 years ago
30cc8b7a92
Merge pull request #15554 from heavengate/yolo_loss_darknet
...
Yolo loss darknet
6 years ago
c4b9eac11a
fix threshold_relu_op ( #15594 )
...
test=develop
6 years ago
16d54f7f23
Return parent_idx in beam_search op ( #15520 )
...
* Refine beam_search_op to output an extra parent_idx tensor.
test=develop
* Fix the unittest test_beam_search_op.
test=develop
* Fix the merging mistake.
test=develop
6 years ago
72ee3c6232
Merge pull request #15398 from jerrywgz/add_axis_for_boxcoder
...
Add axis for boxcoder
6 years ago
e402c0ec7d
test=develop
6 years ago
334f697da9
test=develop
6 years ago
d3eeb92bba
Merge pull request #15491 from tink2123/new_align_corners
...
add align_corners and align_mode for image_resize
6 years ago
3046799ecd
Merge branch 'develop' into add_clip_op
6 years ago
733bb82ec0
downsample -> downsample_ratio. test=develop
6 years ago
577424e5ec
use darknet loss and trick
6 years ago
af124dcdf6
fix API error
6 years ago
20200e126d
fix some typo test=develop
6 years ago
2b89f59055
add attr use_label_smooth test=develop
6 years ago
8218e30176
add gtscore. test=develop
6 years ago
32d533c2cd
cache obj_mask and gt_match_mask. test=develop
6 years ago
e7e4f084e5
ignore pred overlap gt > 0.7. test=develop
6 years ago
db8ff57a61
remove useless code and update doc. test=develop
6 years ago
0c4acc8305
imporve yolo loss implement. test=develop
6 years ago
192d293854
use stable Sigmoid Cross Entropy implement. test=develop
6 years ago
7bc8481c62
Merge pull request #15418 from jerrywgz/refine_nms
...
Refine nms
6 years ago
aaf756272f
remove inplace arg, test=develop
6 years ago
a0c63f1106
add align_flag
...
test=develop
6 years ago
88bd7e1a61
Merge pull request #15027 from shippingwang/shufflechannel
...
Add Shuffle Channel Operator
6 years ago
b64cdaf6dc
modified default parameters
...
test=develop
6 years ago
466a10dcdd
refine code, test=develop
6 years ago
c58555067e
Merge pull request #14731 from jacquesqiao/optimize-cpp-reader
...
Optimize cpp reader
6 years ago
a39240c3b6
add attr variance for box coder, test=develop
6 years ago
db9e700ba1
default use pin place && test=develop
6 years ago
3008fa1261
Add the CUDA kernel for beam_search op ( #15020 )
...
* Refine the beam_search op and test.
* A basic CUDA implementation of beam_search for small batch_size.
* Implement CUDA kernel for beam_search_op.
* Use multiple CUDA threads in the same block to select the top beam.
* Update the python api of beam_search op.
* Enable extend function in CPU kernel of beam_search op.
* Unify the CUDA codes.
test=develop
* Unify the CPU kernel of beam_search op.
* Ensure the seletced items of beam_search_op's CPU kernel sorted by scores.
* Update the description of beam_search in API.spec.
* Enable the use of CUDA kernel in beam_search op.
* Exclude the beam_search's CUDA unittest when there is no CUDA gpu, and delete some debuging statements.
test=develop
* Follow comments.
test=develop
* Call the CPU kernel for beam_search op when batch_size > 4.
test=develop
* Remove the except of is_empty op in PrepareData.
test=develop
6 years ago
78145c7dff
modified some comments
...
test=develop
6 years ago
5cfc40dea8
nce add check sample lables, test=develop ( #15463 )
...
* nce add check sample lables, test=develop
6 years ago
e448bdb298
modified some comments
...
test=develop
6 years ago
88744e4ab8
fixed some errors
...
test=develop
6 years ago
6dfd789bfc
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_nms
6 years ago
6928f8318f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_axis_for_boxcoder
6 years ago
e60c8438fc
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_clip_op
6 years ago
58cb18d9d9
Merge pull request #15322 from velconia/imperative_resnet
...
Imperative Resnet
6 years ago
48cc484643
add align_corners and align_mode for image_resize
...
test=develop
6 years ago
d243e555eb
Merge pull request #15080 from jacquesqiao/optimize-assign
...
Optimize assign
6 years ago
11f1baa406
refine code, test=develop
6 years ago
Dun
Tao Luo
sneaxiy
Dang Qingqing
kolinwei
jerrywgz
Xin Pan
chengduo
Yiqun Liu
Kaipeng Deng
tink2123
ruri
乔龙飞 Qiao Longfei
tangwei12