hedaoyuan
|
773d064ae0
|
Use MinSizeRel compile third_party library when build for mobile inference.
|
8 years ago |
xzl
|
07e5d86787
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into protobuf_lite
|
8 years ago |
|
xzl
|
54a03ab31d
|
add paddle_protobuf_generate_cpp() Func which could dynamic set RUNTIME_LITE mode for paddle
|
8 years ago |
Yan Chunwei
|
843ed8e320
|
dynamic recurrent op forward c++ implentation (#4597)
|
8 years ago |
|
hedaoyuan
|
aa1e92ec9a
|
Merge branch 'develop' of https://github.com/baidu/Paddle into inference
|
8 years ago |
Yi Wang
|
f985700abf
|
Resolve conflict
|
8 years ago |
Yu Yang
|
84500f9487
|
Change `PADDLE_ONLY_CPU` to `PADDLE_WITH_GPU`
By shell command
```bash
sed -i 's#ifdef PADDLE_ONLY_CPU#ifndef PADDLE_WITH_GPU#g' `find ./paddle/ -name '*.h' -o -name '*.cc' -o -name '*.cpp' -o -name '*.c' -o -name '*.cu'`
sed -i 's#ifndef PADDLE_ONLY_CPU#ifdef PADDLE_WITH_GPU#g' `find ./paddle/ -name '*.h' -o -name '*.cc' -o -name '*.cpp' -o -name '*.c' -o -name '*.cu'`
```
|
8 years ago |
|
Yi Wang
|
5f51d0afc4
|
Add -D PADDLE_WITH_CUDA in cmake/configure.cmake
|
8 years ago |
|
hedaoyuan
|
d94dd75e3b
|
Remove string, scripts and retain cuda.
|
8 years ago |
|
hedaoyuan
|
ea4672bea0
|
Remove optimizer.
|
8 years ago |
|
hedaoyuan
|
33299ef972
|
Remove cuda.
|
8 years ago |
|
hedaoyuan
|
3c66b307f7
|
Remove the pserver, trainer, evaluators and some useless gradientmachines when compile mobile inference library.
|
8 years ago |
Qiao Longfei
|
d0ad82cff1
|
fix nv_library (#4370)
* fix nv_library
* fix symbol in gpu_info.h
|
8 years ago |
Tao Luo
|
330e9929ec
|
Merge pull request #4310 from tensor-tang/vgg
enable VGG with MKLDNN layers
|
8 years ago |
Yiqun Liu
|
2d8467ee9d
|
Merge pull request #4320 from Xreki/fix_android_linking_error
Fix bug in cc_library, when merging several libraries into one on Linux.
|
8 years ago |
|
Liu Yiqun
|
dd2f477a50
|
Fix bug in cc_library, when merging several libraries into one on Linux.
|
8 years ago |
tensor-tang
|
f96d31d552
|
only link iomp when with MKLDNN and MKLML
|
8 years ago |
|
Liu Yiqun
|
35dd981db0
|
Fix linking error on mac.
|
8 years ago |
|
Yiqun Liu
|
a9202e89cf
|
Merge pull request #3764 from Xreki/build_ios
Add the support of the cross-compiling for iOS devices
|
8 years ago |
|
Liu Yiqun
|
46ee8ceecb
|
Move the sreach and link of veclib to iOS's toolchain file .
|
8 years ago |
|
Luo Tao
|
70cadf9350
|
Merge branch 'develop' into use_op
|
8 years ago |
|
Luo Tao
|
0597f4de99
|
auto generate USE_OP() in pybind.cc
|
8 years ago |
|
Liu Yiqun
|
76338fb6a7
|
Merge branch 'develop' into build_ios
|
8 years ago |
|
Liu Yiqun
|
9e524fce1b
|
Remove the generation of executable, such as paddle_trainer, when setting WITH_C_API.
|
8 years ago |
|
Liu Yiqun
|
b4d710ce12
|
Merge branch 'develop' into build_ios
|
8 years ago |
|
Liu Yiqun
|
5991a35e58
|
Install the automatically built, compiled libraries under third_party when
executing `make install` and WITH_C_API is set.
|
8 years ago |
|
Liu Yiqun
|
7976876993
|
Add the finding of iOS's native vecLib.framework.
|
8 years ago |
|
Liu Yiqun
|
d1f5f49826
|
Merge branch 'develop' into build_ios
|
8 years ago |
|
Liu Yiqun
|
a98c9e6bbf
|
Add third_party_android back to the cache directories to speedup travis.
|
8 years ago |
|
Liu Yiqun
|
2fc012c54c
|
Merge branch 'develop' into build_android_clang
|
8 years ago |
|
Liu Yiqun
|
8e5f54320f
|
Refine the toolchain file of Android to use clang as default compiler.
|
8 years ago |
|
Liu Yiqun
|
50c63dc9e7
|
Merge branch 'develop' into build_android_clang
|
8 years ago |
|
Liu Yiqun
|
0ea4e2588b
|
Update the version of OpenBLAS and set the sysroot to fix compiling error using clang on Mac.
|
8 years ago |
|
Liu Yiqun
|
fb93a8be27
|
Add paddle_pserver back to c-api library, because it is used in Evaluator.h.
|
8 years ago |
|
Liu Yiqun
|
36fc405261
|
Merge branch 'develop' into build_ios
|
8 years ago |
|
Luo Tao
|
2e8d47dd09
|
simplify and make quiet in the download of mklml.cmake
|
8 years ago |
|
Liu Yiqun
|
aeea8ab1c4
|
Remove the linking of train-related libraries when cross-compiling for Android and iOS.
Recover the mistakenly deleted WARPCTC variable in cmake.
|
8 years ago |
|
Liu Yiqun
|
d57ffc4557
|
Deliver the cross-compilng platform-specific args to external libraries.
|
8 years ago |
|
Luo Tao
|
6479118895
|
fix download mklml error
|
8 years ago |
|
Liu Yiqun
|
0cb3906d5c
|
Merge branch 'develop' into build_ios
|
8 years ago |
|
Liu Yiqun
|
473161f919
|
Merge branch 'ios_cross_compile' into build_ios
|
8 years ago |
|
Liu Yiqun
|
224f8b06f4
|
Support building for multiple architecures at one time.
|
8 years ago |
|
Liu Yiqun
|
d2435ba81c
|
Disable the finding of python, swig related packages completely when WITH_PYTHON and WITH_SWIG_PY are set to OFF.
|
8 years ago |
|
Liu Yiqun
|
fb38e6620e
|
Seperate the codes that cannot and don't need to build for iOS devices.
|
8 years ago |
|
Liu Yiqun
|
43dde9353b
|
Refine the cross-compiling toolchain file for iOS.
|
8 years ago |
|
Liu Yiqun
|
f241773c4f
|
Support to use clang for Android cross-compiling.
|
8 years ago |
|
zlx
|
11cb75f8bb
|
add the ios cross compile
|
8 years ago |
|
tensor-tang
|
2efac83aaa
|
Merge remote-tracking branch 'upstream/develop' into merge
|
8 years ago |
hedaoyuan
|
a683a56f69
|
Merge pull request #3549 from hedaoyuan/convolution
Use EigenBlasGemm improve convolution computing performance in ARMv7 environment.
|
8 years ago |
Yi Wang
|
1eb98e2fef
|
Set the default cuDNN installation path
|
8 years ago |
|
tensor-tang
|
462b9b1d20
|
update mkldnn tag v0.10
|
8 years ago |
|
hedaoyuan
|
adcca2cc06
|
Add PADDLE_USE_EIGEN_FOR_BLAS macro.
|
8 years ago |
qijun
|
13c20ad39e
|
remove --default-stream per-thread nvcc flag
|
8 years ago |
|
qijun
|
5e62605c69
|
Merge remote-tracking branch 'baidu/develop' into implement_EigenCudaStreamDevice
|
8 years ago |
|
qijun
|
95fe318e3e
|
init
|
8 years ago |
|
tensor-tang
|
318fee8389
|
refine cblas
|
8 years ago |
|
tensor-tang
|
2be3d32711
|
use shared lib when mkl
|
8 years ago |
Yi Wang
|
2ca13944b6
|
Merge pull request #3378 from gangliao/proj_name
Rename PROJ_ROOT to PADDLE_SOURCE_DIR and PROJ_BINARY_ROOT to PADDLE_…
|
8 years ago |
|
Yiqun Liu
|
d2258a41a7
|
Merge pull request #3400 from Xreki/capi_read_assets
Support to create a gradient machine with merged model in c-api
|
8 years ago |
|
Liu Yiqun
|
9dccdd77a1
|
Add c-api interface, paddle_gradient_machine_create_for_inference_with_parameters,
to create a gradient machine for inference using merged model with parameters
which is genearted by `paddle merge_model`.
|
8 years ago |
|
qiaolongfei
|
3df8ee1ff7
|
use lib-python as python test dir
|
8 years ago |
|
qiaolongfei
|
c304e02813
|
fix py_padde test
|
8 years ago |
liaogang
|
7a56d46a8a
|
Rename PROJ_ROOT to PADDLE_SOURCE_DIR and PROJ_BINARY_ROOT to PADDLE_BINARY_DIR
|
8 years ago |
|
qiaolongfei
|
e2ccbccb02
|
support python test without installation python package
|
8 years ago |
fengjiayi
|
501da57fb8
|
Merge pull request #3316 from wangkuiyi/remove_wget_progress_info
Remove wget progress info from downloading MKLML
|
8 years ago |
|
Yi Wang
|
d777ac52bc
|
Simplify MKLML download command
|
8 years ago |
|
Yi Wang
|
4b321446dc
|
Remove wget progress info from downloading MKLML
|
8 years ago |
|
Yi Wang
|
af2040bc7e
|
Change git tag of any
|
8 years ago |
|
gangliao
|
50fe7abef4
|
Merge pull request #3295 from gangliao/cpplint_new
enhance cpplint cmake
|
8 years ago |
|
Yi Wang
|
953c51f588
|
Merge pull request #3282 from wangkuiyi/eigen_warning
Add --expt-relaxed-constexpr to surppress Eigen warnings
|
8 years ago |
|
gangliao
|
ec2c753c3d
|
Merge pull request #3290 from PaddlePaddle/cblas_new
Import HPC linear algebra libs as cblas target for cc_library deps
|
8 years ago |
|
liaogang
|
3ce497ca5a
|
enhance cpplint cmake
|
8 years ago |
|
Yi Wang
|
498e9de405
|
Merge pull request #3286 from wangkuiyi/fix_any
Use PaddlePaddle fork of any
|
8 years ago |
|
Yi Wang
|
5ceb7d12d6
|
Merge branch 'develop' of https://github.com/paddlepaddle/paddle into eigen_warning
|
8 years ago |
|
liaogang
|
7ecdf6ad93
|
Add py_test
|
8 years ago |
|
qijun
|
b36f3ae761
|
follow comments and add flags in cuda 8.0
|
8 years ago |
|
liaogang
|
2f9c443be3
|
fix name typo
|
8 years ago |
|
liaogang
|
47c0116653
|
Import HPC linear algebra libs as cblas target
|
8 years ago |
|
Tao Luo
|
fd64369f30
|
Merge pull request #3244 from tensor-tang/issue3232
remove global linker and exe from mkldnn iomp
|
8 years ago |
|
Yi Wang
|
f6f312418d
|
Move --expt-relaxed-constexpt as a NVCC flag
|
8 years ago |
|
Yi Wang
|
0c2f472ca4
|
Use PaddlePaddle fork of any
|
8 years ago |
|
Yi Wang
|
6b51540419
|
Add --expt-relaxed-constexpr to surppress Eigen warnings
|
8 years ago |
|
Yi Wang
|
9283233650
|
Mark the reason of using a personal repo temporarily
|
8 years ago |
|
Yi Wang
|
06ab61fd23
|
Use my personal version of gflags to surpress unnecessary warnings
|
8 years ago |
|
Yu Yang
|
a720d21dcd
|
Paddle always not use `virtualenv` to unit-test
* User must install Paddle python package before unittest.
* Or use docker to build Paddle
|
8 years ago |
|
Yu Yang
|
329370e8ca
|
Merge branch 'develop' of github.com:baidu/Paddle into feature/fast_python_unittest
|
8 years ago |
xuwei06
|
3c798828a9
|
SKip cpplint if source is not changed.
|
8 years ago |
|
tensor-tang
|
547fb6fbfc
|
Merge remote-tracking branch 'upstream/develop' into issue3232
|
8 years ago |
|
tensor-tang
|
e71976f221
|
remove global linker and exe from mkldnn iomp
|
8 years ago |
|
liaogang
|
430b421906
|
Fix conflicts
|
8 years ago |
|
liaogang
|
b58725bd51
|
Add cpplint for *.h and cuda *.cu
|
8 years ago |
|
Yu Yang
|
359e79e543
|
Faster Python unittest
* Do not use `virtualenv` in Docker
* Split unittest into many target, make them parallel
|
8 years ago |
|
Tao Luo
|
ca39600930
|
Merge pull request #3168 from tensor-tang/develop
update MKLDNN cmakes
|
8 years ago |
|
tensor-tang
|
e6f62f79c3
|
add meesage and cmake cache arg
|
8 years ago |
|
tensor-tang
|
1bd64f18ed
|
move MKLDNN and MKLML install path to build third party path
and disable both when build doc and MacOS
|
8 years ago |
|
qijun
|
edb57292f0
|
add cmake patch for gcc version larger than 4.9
|
8 years ago |
|
qijun
|
5364b3944e
|
use cuda default stream
|
8 years ago |
|
qijun
|
d962c2a997
|
fix bug in CUDADeviceContext
|
8 years ago |
|
Tao Luo
|
ef28f66f86
|
Merge pull request #2940 from tensor-tang/develop
enable Intel MKLDNN library and Intel MKL small package
|
8 years ago |
|
tensor-tang
|
95ce961cef
|
fix permission issue
|
8 years ago |
