c3ebff5e70
modified demo/gan following emailxuwei comments
8 years ago
9a02bd419c
fixed a small bug in demo/gan/README.md and testMatrix.py
8 years ago
3ebf6aaf9b
fixed a gpu bug in trainer API to train gan using GPU
8 years ago
70b78b74ff
fix a bug in BatchNormBaseLayer.cpp and add a test for it
8 years ago
fec6f8091f
Skeleton for Generative Adverserial Nets
8 years ago
20b416d256
Resolve conflict from git pull upstream develop
8 years ago
09a5b8bd4d
consolidate img_conv.conf in test_NetworkCompare
8 years ago
0e78171802
changed group=2 in test_NetworkCompare
8 years ago
4641285c75
Added a test in test_NetworkCompare to verify cudnn and exconv, fixed a bug in cudnn_conv in dealing with groups
8 years ago
ba786b2e9e
only do test_ConvUnify when there is a gpu
8 years ago
60bf1a476a
Modifed ExpandConvBaseLayer to unify paras between expand and cudnn
8 years ago
fd92594353
added test_ConvUnify
8 years ago
c410382145
add some comments
8 years ago
1df826e767
Add a AutoCompare and move some test form test_matrixCompare.cpp to test_Matrix.cpp
8 years ago
f340f37f02
Change atomicAdd to paddleAtomicAdd
8 years ago
a816443e11
Add submodule warp-ctc.
8 years ago
1d1a04c033
follow comments on rectangle CNN
8 years ago
5a97c98d6c
Merge branch 'develop' into warpctc
8 years ago
d114d8976a
Merge branch 'develop' into conv
8 years ago
31e57175d9
Merge pull request #655 from PaddlePaddle/release/v0.9.0
...
Release/v0.9.0
8 years ago
f70fc4a439
move some test from test_matrixCompare.cpp to test_BaseMatrix.cpp and test_Matrix.cpp
8 years ago
18645134bd
Follow comments
8 years ago
ff5ca6927f
Merge branch 'develop' of github.com:baidu/Paddle into feature/refine_doc_drnn
8 years ago
4d487c6f35
Integrate warp-ctc as WarpCTCLayer, including unitest and layer interface.
8 years ago
765735b698
Merge pull request #645 from reyoung/feature/improve_paddle_version_check
...
Add `set -e` for paddle boot up script.
8 years ago
1d8d957322
Add `set -e` for paddle boot up script.
...
* error when paddle has a wrong version number.
8 years ago
5de5453d15
add code comments for deepSwap
8 years ago
c7f96de12e
add unit test for deepSwap
8 years ago
cf205d0d43
deepSwap
8 years ago
1873945dc7
Add test_Matrix.cpp for auto compare member functions of class Matrix
8 years ago
1bac8e60c3
modify the file name test_matrix.cpp to test_SparseMatrix.cpp
8 years ago
409a8a181d
Merge branch 'develop' of https://github.com/baidu/Paddle into auto_compare
8 years ago
85e0cd709c
move some BaseMatrix test from test_matrixCompare.cpp to test_BaseMatrix.cpp
8 years ago
1f743d381c
Redesign test_period meaning:
...
* always do test on all test data
* do test at the end of each pass if test_period=0, otherwise do test if test_period batches passed
8 years ago
ff7b4284f4
Add a auto compare for BaseMatrix
8 years ago
96615fe329
merge develop, fix conflict
8 years ago
a3a7e76c7f
Respond to Helin's comments
8 years ago
88802d15cd
Remove template execution results from Dockerfile.m4
8 years ago
aca08255d2
Allow docker build to build from local Git commit.
8 years ago
a7671dee47
refine build_docker_image.rst
8 years ago
341688b583
Bumping up version number
8 years ago
6aece5060b
Stash
8 years ago
da7c0f1326
Format sequence_nest_rnn_multi_unequalength*.conf
8 years ago
e488001675
Merge conflict with hl_cuda_device.cc
8 years ago
514cbeff09
Merge branch 'develop' of github.com:baidu/Paddle into feature/refine_doc_drnn
8 years ago
9ea0661a82
clang format off on some cuda .cc file
8 years ago
80c68d38ff
clang format .cc .h .cpp .c and .hpp file
8 years ago
e9f50bd50b
Merge branch 'develop' into feature/add_clang_format_plugin
...
* Merge conflict using theirs
8 years ago
731fe950c4
Change auto => size_t in BaseMatrix.cu
...
* Because it is a cuda source file, and we need to support c++ 03 in
cuda.
8 years ago
d42fbed02d
Fix several cpp issues
...
* Different Type compare.
* ostream << should pass a const object.
* remove always true checks.
8 years ago
ccea3b026e
Add style check for *.cc files in cuda directory
8 years ago
049f9d3a1c
Fix a pointer comparison bug in hl_dso_loader.cc
8 years ago
5fdaa966ee
Merge pull request #541 from reyoung/feature/fix_diff_type_compare
...
Fix several cpp issues
8 years ago
5e738ca333
Change auto => size_t in BaseMatrix.cu
...
* Because it is a cuda source file, and we need to support c++ 03 in
cuda.
8 years ago
8393c19ccf
Add recursive mutex and counter for gpu profiler
8 years ago
20600e7036
Fix several cpp issues
...
* Different Type compare.
* ostream << should pass a const object.
* remove always true checks.
8 years ago
0ed23581f1
Merge branch 'develop' of github.com:baidu/Paddle into feature/refine_doc_drnn
8 years ago
a49d1d9529
Refine the original hrnn documentation.
8 years ago
20aac5bba1
Add style check for *.cc files in cuda directory
8 years ago
5a67a04e63
Merge pull request #511 from baidu/release/v0.9.0
...
Sync Release Branch => Develop
8 years ago
7b1d3c77c8
Refine ver2num function, add comments
8 years ago
5b1ba87320
Add version check for paddle
8 years ago
f28f2e0ab7
Merge branch 'develop' of https://github.com/baidu/Paddle into profiler
8 years ago
2c84c1ecfb
Add profiler object and update docs
8 years ago
496d64ebdb
Support rectangle input for CNN
8 years ago
65612425f2
Merge pull request #384 from wangyang59/trainerAPIGpu
...
Modified API to use FLAGS_use_gpu as useGpu default value
8 years ago
84cab2c763
Merge conflict with develop branch
8 years ago
e8c0fb9e14
Add GPU Profiler unit test
8 years ago
2e9ea1cece
Add Gpu profiler interface
8 years ago
4705083ee1
Update dockerfile tags
8 years ago
c60f57b6e3
Fix some problems in Debian build scripts.
...
* Mount local Paddle instead of git clone from remote.
* Use official chinese ubuntu source instead of 163 mirror.
8 years ago
f702484ee2
Fix forwardTest for ids in python swig.
...
* unittest need to be added. But fix the bugs first.
8 years ago
9115ab1c11
Merge pull request #450 from reyoung/feature/pre-commit-hooks-scripts
...
Feature/pre commit hooks scripts
8 years ago
4c86285a99
modifed Paddle.swig to specially handle UnsupportError only
8 years ago
b207535198
Add setUseGpu in PaddleAPI.h and handle UnsupportedError in swig with meaningful message displayed
8 years ago
91e6dcb68f
fixed a bug in Paddle::Vector::createCpuVectorFromNumpy
8 years ago
70fecee080
add unittest for Matrix and Vector in API
8 years ago
f22573bdaf
changed to isUsingGpu() in PaddleAPI.h and throw exceptions instead of CHECK
8 years ago
2be3a74779
Modified API to use FLAGS_use_gpu as useGpu default value
8 years ago
a6ad9a1608
Fix unittest
...
Change-Id: Ic80845c892c96c37a0df0ddc433fe1aeaa5a9d1c
8 years ago
bf6f690f31
Add ScalingProjection
...
out = w * input
where w is a parameter of size 1
Change-Id: Ife682d62323ceb1a20cbbf6269421b20a862d888
8 years ago
48ea047100
Bug fix in testing mode.
8 years ago
836d61382f
Update pre-commit-config
...
* Check all files by pre commit hooks
8 years ago
b87427b682
Merge pull request #445 from QiJune/format_py_code_1st
...
format python code in demo, doc, doc_cn and paddle directories
8 years ago
5591292b7a
modifications according to comments
8 years ago
319742c641
format python code in demo, doc, doc_cn and paddle directories
8 years ago
728defbec9
copy the data when createSparseMatrix
8 years ago
069d0004dc
multi_binary_cross_entropy when ids vector is provided
8 years ago
818c81fecc
Merge pull request #428 from reyoung/feature/add_check_env_for_docker
...
Add checkout name for Dockerfile
8 years ago
ca0bb40c6e
Merge pull request #300 from QiJune/feature/sppnet
...
add SpatialPyramidPoolLayer c++ support
8 years ago
ebb153b067
Merge pull request #416 from yu239/hl_activetype
...
change the act.name for LinearActivation() to "linear" so that it won't fail in hl_activetype; also fix the hasinputsset in submodel
8 years ago
eb3bf9ea64
Merge pull request #360 from emailweixu/sum_cost
...
Add SumCost
8 years ago
45f6e1abee
change hlactivetype instead of act.name
8 years ago
64b7561ce6
Add checkout name for Dockerfile
...
* Because in dockerhub, we cannot set the `docker build `running
directory, we could only use `git clone` command to get the latest
code if we put `Dockerfile` in subdirectory
* But the `git clone` will checkout the default branch only, so here
we add a `ENV` in Dockerfile to checkout special branch or tag in
git repo. We could change it to `V0.9.0` tag when it release.
8 years ago
9dd588b414
fix merge conflicts
8 years ago
8295eb91bf
Merge pull request #287 from gangliao/bilinear
...
Add bilinear interpolation layer
8 years ago
cfc965d52c
Merge pull request #269 from wangyang59/deconv
8 years ago
f27ff4d8a4
Revise code
8 years ago
61444d903d
Merge remote-tracking branch 'baidu/develop' into feature/sppnet
8 years ago
5ccf84ab2e
Merge pull request #383 from lzhao4ever/fix_matrix_inverse
...
Fix matrix inverse unittest to be more robust
8 years ago
af7a50c0d4
minor changes on deconv implementation and add protostr test for deconv layer
8 years ago
449120991e
minor change to convTransLayer test in test_LayerGrad
8 years ago
53e1629a43
Refactored imageSize in ConvBaseLayer to MathUtil
8 years ago
03f4b1d4d2
minor changes on deconv per luotao1 comments
8 years ago
7a322df0a8
deconv implementation mionr changes in ConvBaseLayer.cpp and config_parser.py
8 years ago
fb20187aaa
deconv layer implementation modification following luotao1 comments
8 years ago
3d72e94939
rebase deconv implementation with develop branch and resolve conflicts with pull#218 commit 45c81a414f
8 years ago
5e4cc241ac
Revised deconv implementations according to luotao1
8 years ago
5fff96f532
add another small test in test_LayerGrad for convTransLayer
8 years ago
e68b50ad09
fixed a bug in refactoring ExpandConv/TransLayer
8 years ago
2575b74fee
refactored ExpandConvLayer and ExpandConvTransLayer with ConvBaseLayerCpu
8 years ago
aa2cd2ce8f
Refactor ExpandConvTransLayer to share codes with ExpandConvLayer
8 years ago
bda259bb18
added more test on convTrans layer and comments
8 years ago
70e44732c2
added convTrans test and python components
8 years ago
5c88f07262
initial take on deconv layers
8 years ago
b62c80f156
qfg
8 years ago
65b8bb2583
Fix a bug
8 years ago
1c9f6f7f99
Update
8 years ago
f173341fb2
Merge remote-tracking branch 'baidu/develop' into feature/sppnet
8 years ago
0feecbd13c
modify on docs
8 years ago
dfbde28ad3
add some code comments for SppLayer
8 years ago
0a0c55d228
more friendly test options
8 years ago
0519cc6423
Merge branch 'develop' of https://github.com/baidu/Paddle into bilinear
8 years ago
c6a0298e2a
create PR to polish test_period meaning
8 years ago
e6c83f4ec0
some tiny fixs ( #406 )
...
* some tiny fixs
* use VLOG(3)
8 years ago
eaf3dec9c5
follow comments
8 years ago
bc2b521c24
Follow comments
8 years ago
992ac8f9a1
Implement setDiag() with BaseMatrix::assign()
8 years ago
8c40bfd0fc
Make matrix well-conditioned when unittest inverse
8 years ago
70e04683dd
add getSize method for PoolProjection
8 years ago
3553576e6e
Merge remote-tracking branch 'baidu/develop' into feature/sppnet
8 years ago
e2c0713589
follow comments
8 years ago
cc04a7d7ab
Merge branch 'develop' of https://github.com/baidu/Paddle into bilinear
8 years ago
db1757556e
Follow comments
8 years ago
125c19a3c3
fix some nvcc compile options ( #392 )
8 years ago
57bc6238d9
enable swig unittest in travis-ci ( #394 )
...
* Init
* Add numpy deps
* Refine
8 years ago
db569f293e
fix merge conflict
8 years ago
f06f4dfbae
py_paddle link zlib( #393 )
8 years ago
a07da94939
fix floating-point overflow problem of tanh ( #355 )
8 years ago
5ece5c96ad
add python wrap for sppLayer
8 years ago
38764bf908
Add sum_cost to document
...
And rebase
Change-Id: I7ea234b3aa8fc70675af15d91db08242c43fb5ff
8 years ago
ebad8e525d
Add SumCost
...
This allows user to implement any type of cost by summing over the output of non-cost layers.
Change-Id: Ic55aaabbf0c1299e70b8e48a0effcc91f8f5bd29
8 years ago
4905751a22
Add define for double getrf, getri ( #381 )
8 years ago
5a1e7dbc6a
Fix conflict
8 years ago
e802471c58
abstract outputSize function in CNN-related layers ( #314 )
8 years ago
724d6dd40a
Replace outputH to batchSize
8 years ago
d373c10091
Fix bilinear interp bug
8 years ago
45b8c47e04
Add img_size for unit test
8 years ago
bd38facada
Fix conflict
8 years ago
57348806b5
Follow comments
8 years ago
e05f4ff267
Fix SRL hang when exit. ( #291 )
...
* Fix SRL hang when exit.
* Error occurred when enable Async Load in TestDataProvider.
* It because DataProvider is calling getNextBatchInternal in one thread, and destructing DataProvider in other thread.
* Add wait routine in DataProvider destructing.
* Also fix another bug, when destructing TestDataProvider and do not read any test data.
Fix #286
* Follow comments, Use mutex is cool!
8 years ago
36bda94eb4
include mkl_lapacke.h ( #359 )
8 years ago
6c3a678c9a
Add elementwise math operations ( #343 )
...
* Add elementwise math operations
This allows use to use expressions like: y=log(1+exp(x))
Also added unittests for ActivationFunction
* Enforce keyword arguments for non-positional arguments
* Add LogActivation to doc
8 years ago
3e2dc77cc7
Add code coverage and coveralls ( #296 )
8 years ago
33004ecfb7
Fix glog check type unmatch in Util.cpp ( #353 )
...
* Fix glog check type unmatch in Util.cpp
#352
8 years ago
3424a4c0d8
Fix bug and redundant code in hl_dso_loader.cc ( #306 )
8 years ago
1de75c039f
report error when use parallel_nn to train recurrent_nn model ( #335 )
8 years ago
5f2059db05
Add matrix inverse ( #240 )
...
* Add matrix inverse
8 years ago
9f9b4afcdb
install the right python package version ( #326 )
...
For multiple installation of paddle, there might be multiple versions of python package at opt/paddle/share/wheels/. We should install the right version.
Ideally, we should remove the wrong versions when install. But it's not easy to do this with cmake.
Change-Id: Ida8a8d60643ad9e42cf1c85776de9122d5ba1392
8 years ago
b282caf4f6
Merge remote-tracking branch 'baidu/develop' into feature/sppnet
8 years ago
968464cc60
Fix a bug in testOnePeriod. ( #322 )
...
* Forget to finishTestPeriod in testOnePeriod.
* Fix #318
8 years ago
5acf136615
Bug fix in CudnnConvLayer, which will lead to destruction error. ( #317 )
8 years ago
ee028bb5ea
Add How to build docs ( #312 )
8 years ago
766a61c374
fix conflict with baidu/develop
8 years ago
fcf177fc4b
reuse code of PoolProjection in PoolProjectionLayer
8 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.
8 years ago
cdac60f616
add SpatialPyramidPoolLayer c++ support
8 years ago
6467c38202
Add default cuda system path ( #192 )
...
* DYLD_LIBRARY_PATH is disable after Mac OS X 10.11
* fix clang + gpu compile error on Mac OS
* fix some words and errors in build docs
8 years ago
460320a41c
fix type unmatch on gcc
8 years ago
fd4eeaf59c
Merge conflict with maxout layer
8 years ago
ddfff3a7fd
Add bilinear interpolation layer
8 years ago
fa24cbdbe1
Support empty Param Block in ParameterSever ( #244 )
...
* Because in cluster maybe use a lot machine to train a model, and some parameter size could be too small for ParameterServer. Then some of pservers could not have any ParamBlock.
* Also, because ports_num or ports_num_for_sparse is too large, then give a warning in runtime.
8 years ago
ca5a5ec480
Make Paddle --save_dir support a directory name ( #277 )
...
* Also fix #243
8 years ago
0e1a22d0fc
set test_period default value to 0 ( #279 )
8 years ago
cbe734b396
Python trainer api ( #193 )
...
* Python trainer API and demo
* Adding missing PaddleAPIPrivate.h
* Adding api_train.sh
* More comments
* Bump up patch version to 0b3
8 years ago
46bd5f53e3
add input sparse data check for sparse layer at runtime ( #247 )
...
* add input sparse data check for sparse layer at runtime,
to avoid invalid data access at pserver end while doing prefetch
* remote sparse design support binary sparse and float saprse both
9 years ago
9c5c38fa2a
FIx check type unmatch in MaxOutLayer ( #242 )
...
Compiled failed on gcc 4.6
9 years ago
07b2e5d54c
Fix dataprovider converter for sparse data
9 years ago
3dd8c9bea4
add maxout layer, including interface and unittest ( #229 )
...
* add maxout layer, including interface and unittest
* follow maxout comments
* auto setting channels
* fix unittest bug in test_RecurrentGradientMachine
9 years ago
e20ff3b3e9
fix build bug in gcc46 ( #236 )
9 years ago
e1f57bfd66
add base class for seqlastin/max/average layer ( #187 )
9 years ago
6bef8390b9
fix bug in some different python environment ( #220 )
9 years ago
58f896c3f4
Speed up PyDP2, support numpy.float array ( #207 )
9 years ago
28bc05b126
Fix sparse training for trainer_count=1 ( #204 )
...
* Fix sparse training for trainer_count=1
For trainer_count=1, the gradient machine is NeuralNetwork, which does not create parameter buf for PARAMETER_GRADIENT for sparse update in Parameter::enableType. But gradient parameter buf is still used in SgdThreadUpdater.
* Minor update to comment
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
4e43a59a18
remove unmerged internal documents ( #205 )
9 years ago
c13bdb15cd
remove redundant HPPL_TYPE_DOUBLE ( #200 )
9 years ago
91df606280
remove some copyfrom in AgentLayer and ExpandLayer, fix warning in seq2seq config ( #183 )
9 years ago
cebdb66768
hierarchical rnn document, add new config example ( #106 )
...
* hierarchical rnn document, add new config example
* update inputs_type of label
* add check for unsupported config
* refine hierarchical document
* refine doc title
* update docs, fix paddle to PaddlePaddle
* follow comments
9 years ago
c2d418dbfd
Merge internal changes ( #198 )
...
* fix DataProvider create function args bug
Change-Id: I9e3a1c535c805bf30204a14aea8d5143ff534784
* remove PserverForPython.h which is not used
Change-Id: I2b27f1f3c11a42766a92fc689f0f5f1f73ee1d70
* add internal document script
Change-Id: Ia0fec79456caea0b271f9903cc13e8a3d32e0774
9 years ago
43f7d7b768
add interface and unittest for nce layer ( #180 )
...
* add interface and unittest for nce layer
* follow comments
9 years ago
3edc4202b0
Add travis for osx ( #189 )
9 years ago
9f244e4a39
Should not compile the two files if -DWITH_AVX=OFF. ( #163 )
...
* If cmake -DWITH_AVX=OFF during configuration, should not compile the file src/hl_math.cc and src/hl_avx_functions.cc.
9 years ago
9e11ca8096
Use C++ 11 atomic_flag in MacOS as spin lock ( #175 )
...
* Use C++ 11 atomic_flag in MacOS as spin lock
* Add unittest for it.
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
0ab332242f
Support MAC OS Sierra ( #169 )
9 years ago
1c2ebe467b
add interface and test of RecurrentGradientMachine ( #156 )
...
* add interface and unittest of RecurrentGradientMachine for the function of multiple Subsequence inlinks with unequal token length
9 years ago
b52039bd11
some bug fix for sparse matrix ( #133 )
...
* some bug fix for sparse matrix
* a minor bug fix
9 years ago
0276f15a45
add scripts to build ubuntu install package. ( #132 )
...
* also refine install docs, too
9 years ago
2920b6bc0d
Merge pull request #138 from gangliao/master
...
Add Mac OS X port
9 years ago
8ddc5faac1
Update Mac OS X port
...
* follow comments to fix bugs
9 years ago
cbb904356c
Remove main function in some unittest.
9 years ago
1fc4352754
refine sparse momentum api and unittest ( #126 )
...
* refine sparse momentum api and unittest
* fix unittests bug
9 years ago
6decbdf4f5
Fix potential dead lock in PyDataProvider2 ( #140 )
...
This bug occasionally causes dead lock in test_RecurrentGradientMachine
In general, conditional_variable::notify should be used together with mutex for changing condition.
9 years ago
4615c5172c
beam search api and unitest in hierarchical rnn ( #122 )
9 years ago
baaaa0b09d
Add NOAVX Docker image scripts. ( #129 )
...
* Add noavx to docker
* Not to use directory to split Docker image.
9 years ago
4bc4a44ed9
Check if paddle is installed when unittest. ( #134 )
9 years ago
0072ef50bf
Fix compile check type failed in linux
9 years ago
23e47bb600
Merge remote-tracking branch 'upstream/master'
9 years ago
a8df411192
Replace random_shuffle using shuffle.
...
* reduce trainer count for unit test on MAC OSX
9 years ago
b15a4783cb
Correctly handling multiple inputs and integer inputs for recurrent_g… ( #114 )
...
* Correctly handling multiple inputs and integer inputs for recurrent_group
* Fix ScatterAgentLayer for generation
* Revert sequence_(nest)_rnn.conf
9 years ago
ffc341675d
Add parallel_nn api and unittest ( #110 )
...
* Add `device` parameter to ExtraAttr in trainer_config_helpers.
* add unittest for it.
9 years ago
1d4bc47805
support gettid() on MAC OS X
9 years ago
2289c141c2
Refine comment for CRF related headers. ( #117 )
9 years ago
77ed98d1a8
fix bugs under kSgdSparseCpuTraining mode ( #100 )
...
Local training with "sparse_update = True" parameter triggers kSgdSparseCpuTraining mode, fix bugs under it.
9 years ago
5bca34ed24
Merge remote-tracking branch 'upstream/master'
9 years ago
32b5557329
Add thread Barrier unit test
9 years ago
c1c07bbef6
Update CudnnConvLayer.cpp
9 years ago
95da095d10
fix cudnn conv bug which occurs in image classfication demo in GTX GPU
9 years ago
7eb29f264b
Try to fix MultinomialSampler ( #102 )
...
* Also refine unittest to multiple iteration to prevent luckily random number.
9 years ago
b8d26ff4d7
fix float except bugs
9 years ago
aaed5cfccc
revert real into float for swig API
9 years ago
7ff8e76229
Shrink batch size on unit test for Mac OS X
9 years ago
8b1c76af3c
comment LOG(info) in createPythonClass
...
* it makes unit test failed.
9 years ago
536bf7d87f
fix poolAllocator unordered_map bug on Mac OS X
...
* using map to replace unordered_map on Mac
9 years ago
9795e01639
Merge remote-tracking branch 'upstream/master'
9 years ago
2c5a6ac095
Optional fields to shrink generated proto size ( #93 )
...
* remove unnecessary field set in ParameterConfig, Evaluators, etc
9 years ago
d2e1b46f24
update beam_search and seqToseq config, and add ExpActivation api
9 years ago
90b9cba7fe
Add min_pool_size, Add default value of should_shuffle ( #70 )
...
* min_pool_size would be infinite by default.
* add unittest for min_pool_size
* Fix bug in can_over_batch_size
* add unittest for can_over_batch_size
* Add DEFINE_PROVIDER_EX
* Add default value of should_shuffle
* When training, the default value of should_shuffle is True.
* When testing, the default value of should_shuffle is False.
* User a set a provider should_shuffle or not by pass it to `@provider`
* should_shuffle can handle a list of value, not just boolean
* Add input order mapping by using name
* Add unittest
* Add check to check input format.
* Default is close for speed reason.
* User could stop train when check error, or continue train without
this train sample.
* use deque instead of vector in generators pool, make erase
generator faster.
* Add chinese/english documentation
* Make should shuffle = false in unittest
* Add python files to depends.
9 years ago
2d13462a2c
Fix incompatible on CUDA atomicAdd operation
9 years ago
4e37b226f4
Revise member variable in private barrier class
9 years ago
75beeaf743
Fix unit test stack trace bug on MAC OS
9 years ago
a6b58cf013
Merge remote-tracking branch 'upstream/master'
9 years ago
92ca98d5f5
Merge remote-tracking branch 'upstream/master'
9 years ago
48eb5ff01f
Fix signed/unsigned comparison for gcc 4.9
9 years ago
05a97ab50e
Further fix the memory for Hierarchical RNN
...
Sequences should be sorted according to the number of subsequences they have.
9 years ago
df28da76c5
try to fix bug for CTCErrorEvaluator.cpp when batch_size > 1 ( #82 )
...
* try to fix bug for ctc_error_evaluator
9 years ago
703cce35a8
Enable docs update. ( #78 )
...
Just forget to add deploy_key to this repository.
9 years ago
7e79c911b4
Temporarily disable automatic doc deployment
...
to restore successful build status. travis DOCS job on master failed due to permisstion issue: https://travis-ci.org/baidu/Paddle/jobs/159851334 @reyoung, please resolve the permission issue.
9 years ago
df0c7cd9c0
Add documentation ci scripts ( #66 )
...
* Add documentation build to ci test
* Push master branch doc changes to gh-pages branch
9 years ago
1f8c21978d
replace fabsf() using std::abs
9 years ago
8a379fce16
add sse instructions support for double on MAC OS
9 years ago
0ba302f7d7
fix bug on paddle api when WITH_DOUBLE
9 years ago
d8366a67fa
fix bug on thread barrier
9 years ago
c7762da330
Merge branch 'master' into merge_icode
9 years ago
9a9de9240d
Correctly handle memory in RecurrentGradientMachine for hirarchical RNN
...
Change-Id: I8e0a8ea6fc2760652d9c76440a539c90860062d3
9 years ago
699d5f2638
modify RecurrentGradientMachine to support unequal length inputs
...
* modify RecurrentGradientMachine to support hasSubSeq sequence inlinks with the same number of sentence but different number of tokens for each sentence
Change-Id: Ic71f00a4bb346b4fa93e650dfb4b1a0d8d2338b0
9 years ago
0f91ea7ebb
use HPPL_STREAM_DEFAULT for layer computation
...
Change-Id: Id66da7b7f5bf9ec80cc19b347e4fb822a5a6f197
9 years ago
688eeefab3
fixed issues with synchronizing streams when copy from gpu to cpu
...
* by default, synchronize default_stream after resizeAndCopyFrom
* add sync in some places after resizeAndCopyFrom using other streams
9 years ago
7481429f99
Change cos_sim to use CosSimLayer layer when size=1 and rename convex_comb_layer to linear_comb_layer
...
Also add an unittest for layers.py, currently for a few things. More need to be added later.
9 years ago
b2d65384e9
Merge remote-tracking branch 'origin/master'
...
# Conflicts:
# paddle/math/tests/test_perturbation.cpp
9 years ago
1df0c7b149
avoid link failed on unit test for mac os
9 years ago
7dbc092c87
fix cudnn version number for batch norm. ( #71 )
...
* fix CUDNN_VERSION for backward of CudnnBatchNormLayer
* fix cudnn version number for BatchNorm
9 years ago
55a1a7588b
Disable a unittest will use large memory by grep
9 years ago
a238b11f83
Fix linux compile
9 years ago
d9870c354f
Merge branch 'master' into mac_port
9 years ago
db13fddcc4
Clean locks.h, Add arch Locks.cpp
9 years ago
eb5cf64b5e
reduce data_layer size of unit test to avoid cuda out of memory on MAC OS
9 years ago
d3eef0c9af
reduce data_layer size of unit test to avoid cuda out of memory on MAC OS
9 years ago
9a9e059717
Merge remote-tracking branch 'upstream/master'
9 years ago
8d0214193e
fix unit test bug when only one gpu
9 years ago
c7ece60e2d
add gettid syscall for MAC OS
9 years ago
50c3dbf6c5
Merge branch 'master' into custom_stack_trace_refines
9 years ago
dcd87fd689
fix CUDNN_VERSION for backward of CudnnBatchNormLayer ( #61 )
9 years ago
0964b7ffc6
Merge branch 'master' into custom_stack_trace_refines
9 years ago
70cceb0cdd
fix compile paddle swig bug on MAC OS
9 years ago
ec2a3eb3c9
Merge remote-tracking branch 'upstream/master'
9 years ago
b664ca0321
auto-tuning SND/REV buff size on MAC OS
9 years ago
87f96f873a
fix dynamic load PaddlePaddle for Mac OS
9 years ago
58f74e2ca1
Add main entry for unit test files and replace memalign by posix_memalign
9 years ago
674d69ce57
Merge pull request #49 from reyoung/LazyInstallPaddleWheels
...
Lazy install Paddle wheels
9 years ago
a3941cbc1f
remove weak attribute for internal FPGA
9 years ago
3f5ce64cab
fix StorageEngine::singleton and std::function<void()> type unmatched.
9 years ago
445d1e8201
Support MAC OS for PaddlePaddle
9 years ago
25fd8731ec
Merge branch 'master' into fix_calc_batch_size_deadlock
9 years ago
0c8aeffb9a
Fix a PyDataProvider2 bug when use calc_batch_size
...
* Need PyGuard when set args for calc_batch_size
9 years ago
62da4ad5b4
Merge branch 'master' into LazyInstallPaddleWheels
9 years ago
b76682c5e3
Exit when pip install failed
9 years ago
1d87c6e0c6
Merge branch 'master' into custom_stack_trace_refines
9 years ago
b159a36e3f
fix redundant macro in hl_device_functions.cuh
9 years ago
6ef0b7ce6d
Merge branch 'master' into LazyInstallPaddleWheels
9 years ago
49a92da24c
Add unittest to CustomStackTrace.
9 years ago
b384af58be
Fix bugs in CustomStackTrace.
...
* Make layer stack trace shows ThreadId, Forward or Backward.
Change-Id: Iba1477adb8c9115c3a67ff2959bb5c878ca706c7
9 years ago
4a880f03c3
Fix bug in NO_AVX when using lstm
...
* libpaddle_cuda add wrong -mavx to some cpp files.
9 years ago
7cf8e0c9e6
Lazy install Paddle wheels
...
* install wheels when invoke paddle script if current python don't
have paddle packages, or installed a previous version.
* Also add `make install` to travis
9 years ago
dbaabc94fb
fix unitest of test_RecurrentGradientMachine, and some tiny doc update
...
Change-Id: I028e402c964ca4f4431cbf8153bea4379dd4df70
9 years ago
903d5c7ec0
bug fix for hl_matrix_classification_error
9 years ago
fdd40e5528
Fix 32-bit gcc compile warnings.
...
Change-Id: Ibc39ca1d1a27d0d28569e29f41a5647659f8c764
9 years ago
7ad55a4e76
Fix ThreadParameterUpdater
...
The reference return type causes ThreadParameterUpdater.cpp:123 seg fault under gcc5.4.
Change-Id: I7a1c155892722076a7cb48793b83d5ee525747d1
9 years ago
721b09eee6
Update Jumbo package to 0.8.0b0.
...
Change-Id: I0b8608feab8f6be5094e8981fc5f65cb401ed415
9 years ago
fbfd24e6d9
revert CRFLayer, remove wrong gpu support
...
Change-Id: I636cf13af5becb1168bc9749266b55580c46f6c9
9 years ago
1e1a33b574
Argument concat for subsequence start positions
...
Change-Id: Ia60c008a8c922f66e6b5e2ca3e488fc4625d6506
9 years ago
c3c76d69a4
Merge branch 'master' of github.com:baidu/Paddle into opensource
9 years ago
daaf5a42ec
Add travis support to PaddlePaddle
9 years ago
d6d85add20
fix bug in findLastSet
...
* findLastSet function: size_t could be uint, ulong, ulonglong
* add default cuda 7.0 in flags.cmake
Change-Id: Ica9d09520e6a1468e5fc97e2c29c1ea7c73dafbb
9 years ago
291318e5a2
delete Bits.h and move its func into utils/Util.h
...
Change-Id: Ib7f7d3c65c9850b8420f9a4799835a7ea03572bd
9 years ago
c974a99325
Remove the extra endif by merge error
...
Change-Id: Ic35a54a94cbd981defff5b44a063dc680cdaaa7c
9 years ago
5de6b97a22
Merge branch 'master' of ssh://gitlab.baidu.com:8022/idl-dl/paddle_opensource_mirror
9 years ago
c9dc794e19
fixed build issue of double definition of atomicAdd on modern GPUs
9 years ago
54d4968d38
Refine code and comments for CRMNormLayer.
...
ISSUE=4612671
git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1501 1ad973e4-5ce8-4261-8a94-b56d1f490c56
9 years ago
wangyang59
xuwei06
Yi Wang
hedaoyuan
liaogang
Liu Yiqun
Luo Tao
qingqing01
Yu Yang
wangkuiyi
xutianbing
wangyanfei01
Yi Wang
backyes
Haonan
qijun
Haonan
emailweixu
Liang Zhao
hedaoyuan
lzhao4ever
Z-TAO
Mark
Zrachel
liuyuan
liaogang
He
Fan Yang
dangqingqing