7be4f2b754
Merge remote-tracking branch 'upstream/develop' into img_separable_conv
7 years ago
94ca8f2152
Make the default layer type for separable conv is exconv
7 years ago
964f01e3e8
fix simple_gru2 doc
7 years ago
126d274ed4
Add separable convolution
7 years ago
56ec40ad9c
Merge pull request #4924 from ranqiu92/attention
...
Add the configuration helper for multi-head attention.
7 years ago
f22402933e
Refine multi_head_attention
7 years ago
81abcdea39
Refine dot_product_attention
7 years ago
06c7686aa1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into attention
8 years ago
e97354152c
unify the bias
8 years ago
1b6a54e286
fix error for annotation
8 years ago
7461b35977
Refine multi-head attention
8 years ago
4ad12a0bd5
Fix bugs of dot-product attention
8 years ago
947c528508
Remove redundant flags
8 years ago
4d15b107f3
Add multi-head attention
8 years ago
78320194aa
refine dot-product attention according to the comments
8 years ago
4545a058bd
add dot-product attention
8 years ago
fc3b129b08
delete the unused comments
8 years ago
bfba756b48
update the version of pre-commit0.13.2
8 years ago
9e74b89877
update notation in networks.py roughly
8 years ago
b90461b9d9
fix the typo of the param description in sequence_conv_pool
8 years ago
34f4f763f9
Update networks.py
8 years ago
f715c740bf
Add_config_parser_for_Conv3D_DeConv3D
8 years ago
a59fa46c1f
Merge branch 'add_paramAttr' of https://github.com/NHZlX/Paddle into add_paramAttr
8 years ago
18f4d24d0d
moidify comment of im_conv_group
8 years ago
5f32cc10c2
Update networks.py
...
modify the format
8 years ago
21a3c9d6f4
add the comments for img_conv_groups
8 years ago
fae3632c91
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_paramAttr
8 years ago
a183a80e2e
add param_attr for img_conv_group(...)
8 years ago
1b6faffeb5
Merge branch 'develop' into fix_recurrent_parse_bug
8 years ago
0ed51ce2e4
fix bug of type check of inputs to recurrent_group.
8 years ago
62da4a1cde
Merge pull request #2802 from reyoung/feature/fix_python_slow
...
Fix slow parsing a recursive depends topology
8 years ago
313e9f551f
Fix slow parsing a recursive depends topology
...
* Fix #2797
* It because trainer_config_helpers' __dfs_travel__ did not record the
node which travelled, and if the topology has a recursive dependency,
there are some nodes will be travelled multiple times.
* Add a `travelled` set to record which node is travelled.
* Also add a unittest for this situation.
8 years ago
68ec558c21
update the md5sum of WMT14 trained model.
8 years ago
5c68aacad1
follow comments.
8 years ago
b0ad9c9074
enable intializing memory state for lstmemory_group.
8 years ago
b34a05d114
Fix travis-ci in new image
...
Fix issue #2562
8 years ago
692d251869
add missing configuration functions in v2 API.
8 years ago
3f1151a54c
Add error clipping to MT demo.
...
* Compose GRU step naive layer in trainer config helpers.
* It is uses mixed_layer for gate.
* It supports ERROR_CLIPPING, DROPOUT
* Add error clipping in MT demo.
* Fix #1143
* Fix #1891
8 years ago
781b85b5fc
rotate_layer and flip_layer * added getMin and getMax for GpuMatrix * gru_step_layer parameter name
8 years ago
c1f9cd9dbe
Merge pull request #1241 from wangyang59/rnnParaShare
...
make gru_group parameters sharable
8 years ago
e1d074abdb
updated comments for gru_group and lstm_group in networks.py
8 years ago
6da7283475
make gru_group parameters sharable
8 years ago
e9549cbb78
Change "Baidu, Inc" into "PaddlePaddle Authors"
8 years ago
a1ba3f442f
format python code in python directory
9 years ago
45c81a414f
Add job=time in trainer, refine cudnn_conv to reduce gpu memory and speed up training. ( #218 )
...
* Add benchmark for PaddlePaddle, tensorflow and caffe
* ConvProjection to reduce memory for goolenet
* Add unit test for ConvProjection.
1. unit test in test_LayerGrad.
2. compare the ConvPorjection and CudnnConvLayer, also compare the concat_layer+img_conv_layer and concat_layer_conv_projection.
* Reduce cudnn_conv memory and add benchmark document.
1. Use TmpMatrix as the workspace in cudnn_conv to reduce gpu memory. It reduce lots of memory.
2. Add benchmark document.
3. fix smallnet_mnist_cifar.py in paddle.
* Add job=time and refine cudnn_conv to reduce gpu memroy and speed up
* Refine cudnn_conv and shared biases operation in concat_layer and mixed_layer.
* follow comments
* follow comments
* Use unique_ptr to prevent memory leaks in CudnnConvLayer.
9 years ago
652b83478f
remove deprecated start input in img_pool_layer ( #237 )
9 years ago
2f82d72ede
Fix bug in yield dictionary in DataProvider. ( #197 )
...
* Fix bug in yield dictionary in DataProvider.
* Also make virtualenv work in Paddle.
9 years ago
e4952ca6ce
Add FAQ ( #128 )
...
* Init commit for doing FAQ
* Add speed up training
* Add graphviz to ci
* Add shared paramter
* Tiny refine
9 years ago
191fafe355
support rectangle padding, stride, window and input for PoolProjection ( #115 )
...
* support rectangle padding, stride, window and input for PoolProjection
* Follow comments.
1. Remove start
2. refine img_pool_a/b.conf for test_NetworkCompare
3. Split unit test
* Modify the test in img_layers.py
9 years ago
d130d18146
Complete unittest for trainer_config_helpers. ( #108 )
...
* Fix lots of trainer_config_helpers bug, and complete unittest for `layers.py`
9 years ago
wangmeng28
qiaolongfei
Cao Ying
ranqiu
peterzhang2029
chengduo
chengduoZH
zlx
caoying03
Yu Yang
Haonan
emailweixu
zhanghaichao
wangyang59
Yi Wang
qijun
qingqing01
luotao1