OrgLion-ML
/
dlib
mirror of https://github.com/davisking/dlib.git
1
0
Fork 0
Code Issues Releases Wiki Activity
6,460 Commits 4 Branches 61 Tags 140 MiB
Commit Graph

6460 Commits

Author SHA1 Message Date
David Miller 600923359f GCC/Clang compatible SIMD code ./dlib/simd/simd**_vec.h (#414) 
* GCC/Clang compatible vector extension SIMD code

* Minimal modifications to dlib for the simd_vec code to work, a few include changes and ifdefs

* Changed tabbing to spaces

* Allow type inference to binary ops on different types of same size

* Added cmake option USE_AUTO_VECTOR, and fixed up preprocessor checks.
It is required to build with gcc/clang auto vectorization

* Changed to intrinsic version due to poor auto vectorization results.
The simd8*_vec are just copies of the C code right now.

* Removed _vec variants, added to existing defines. simd_check.h back in place and removed from dlib/simd.h
 2017-02-01 15:58:40 -05:00
Davis King f6ece5d23a merged 2017-01-30 15:19:16 -05:00
Davis King 4dfeb7e186 Made alias_tensor usable when it is const. 2017-01-30 15:18:39 -05:00
Davis King 7ba8ea9c27 Made dlib::thread_pool use std::thread and join on the threads in 
thread_pool's destructor.  The previous implementation used dlib's global
thread pooling to allocate threads to dlib::thread_pool, however, this
sometimes caused annoying behavior when used as part of a MATLAB mex file.
 2017-01-30 11:03:31 -05:00
Davis King cedccc9fc1 Made global thread pool shutdown not attempted unless the thread pool was 
already enabled.  This also makes the global destruction order a little more
deterministic.
 2017-01-30 10:56:47 -05:00
Davis King cb9cb4e02b Made matrix literal assignment faster for column major matrices. 2017-01-30 09:56:06 -05:00
Davis King bdbf7bb8da Fixed documentation 2017-01-25 06:59:08 -05:00
Davis King 8f4f93b09e merged 2017-01-25 06:56:59 -05:00
Daniel Crispell c15924fa5c add support for loading of RGBA JPEG images (#409) 2017-01-24 19:11:30 -05:00
Davis King 6c6c25a344 merged 2017-01-22 11:32:27 -05:00
RangerUFO 0f60ba365b Fix comments typo in dnn_introduction2_ex (#405) 2017-01-22 10:36:40 -05:00
Davis King aa0fa28533 merged 2017-01-22 10:26:27 -05:00
Davis King 3793e29e0e Added comments about test_one_step() 2017-01-22 10:25:29 -05:00
Davis King c20c11af90 Added test_one_step() to the dnn_trainer. This allows you to do automatic 
early stopping based on observing the loss on held out data.
 2017-01-22 10:25:06 -05:00
Davis King 601d1612e0 Always compile dnn code. 2017-01-21 21:26:15 -05:00
Frank 7ef7ba84b3 Fix imglab readme typo in cd path (#400) 2017-01-16 11:47:06 -05:00
Davis King 5f5684a8fb Made python library and header detection more robust 2017-01-16 06:54:20 -05:00
Davis King 7cf1a9282b merged 2017-01-09 07:15:13 -05:00
Davis King 47e32fb16c merged 2017-01-09 07:14:51 -05:00
Davis King 93610c4e55 Made sure cmake config file pulls in dependent include folders. 2017-01-09 07:14:20 -05:00
Stefan Schweter 45d35e6ce6 Made matrix2 test more robust on ARM64 (#394) 2017-01-08 14:46:35 -05:00
Davis King a4470b0e19 Only turn off CUDA_PROPAGATE_HOST_FLAGS on OS X since that's the only place 
that really needs it off.
 2017-01-06 16:12:54 -05:00
ernestotapiar 365dfd20d1 cmake script to suppress randlib warnings about *.a having no symbols on MacOSX (#390) 
* cmake script to suppress randlib warnings about *.a having no symbols on
MacOSX

* Moved script code into the main CMakeLists.txt file to suppress the
superfluous ranlib warnings all the time.
 2017-01-05 21:04:41 -05:00
Davis King 97151133b0 Changed to avoid linker errors on some compilers. 2017-01-04 17:19:31 -05:00
Davis King fcf211d9c4 merged 2017-01-04 09:10:58 -05:00
Davis King 6a9637ca54 Fixed cmake script so that nvcc gets the correct compiler options sent to it 
(in particular on mac os x, where this flag needs to be set this way for it to
work in some cases).
 2017-01-04 09:10:13 -05:00
Davis King addb19d336 Initialized exception_ptr since some systems, incorrectly, don't initialize it automatically. 2017-01-03 06:35:32 -05:00
rquk 8d76423096 Fixed issue with uninitialised variables in thread_pool_extension. (#386) 
* Fixed issue with uninitialised variables. There are 2 places where std::exception_ptr eptr is not initialised.

* running_gradient.h needs to qualify erfc with 'std::' to avoid compilation error. 
Found when compiling dnn_mmod_face_detection_ex.cpp with RadStudio and Clang compiler.
 2017-01-02 13:08:03 -05:00
Davis King 4a95f8bd72 Made pixel_traits use constexpr everywhere to avoid compiler errors in newer compilers. 2016-12-27 09:33:41 -05:00
Davis King de13fdeb5a Made setup.py always set PYTHON_INCLUDE_DIR and PYTHON_LIBRARY based on 
distutils environment variables for all versions of python, not just conda.
 2016-12-27 09:21:38 -05:00
Davis King d0726637f5 merged 2016-12-27 08:25:13 -05:00
hiiwave c22bedc1c6 Modify setup.py so that it can be installed under conda environment (#377) 2016-12-27 08:18:49 -05:00
Davis King b93b6f36a4 Made the dlib shared library build dependent on the static library build. This 
is because some build systems seem to mess up the parallel build of these two
things and sometimes rebuild and potentially clobber object files during the
parallel build.  I haven't seen this result in a problem, but the race
condition does seem to exist in some build systems.  So this should force the
two top level targets to be build sequentially rather than in parallel, thus
avoiding any possible race condition.
 2016-12-26 08:56:40 -05:00
Davis King 7bf01102d5 merged 2016-12-25 14:41:23 -05:00
Davis King fdbb3b1212 fixed spelling error in comment 2016-12-25 12:32:48 -05:00
Stefan Schweter 2f91cd6c48 Added new test case for split method (unicode non-breaking space) (#374) 2016-12-23 20:09:12 -05:00
Davis King 1cdbcb5a10 merged 2016-12-23 13:02:13 -05:00
Davis King bbe8b12d69 Added matlab_object to the mex wrapper. Now you can have parameters that are 
arbitrary matlab objects.
 2016-12-23 10:12:12 -05:00
Davis King 9beac18b71 Removed C++11 check here since it's now redundant with the checks in algs.h. Also, the check 
doesn't work correctly in visual studio since visual studio doesn't define __cplusplus
to 201103 even though the current version ostensibly supports C++11.
 2016-12-22 20:43:50 -05:00
Juha Reunanen 88ab07db68 Minor documentation fixes (#368) 2016-12-20 06:35:13 -05:00
Davis E. King d994054462 Merge pull request #365 from reunanen/fix-363 
Fix #363 by not setting CUDA_HOST_COMPILER when targeting Visual Studio
 2016-12-19 15:14:49 -05:00
Juha Reunanen 6aa96fdebe Set CUDA_HOST_COMPILER when not targeting Visual Studio 2016-12-19 19:02:27 +02:00
Davis E. King 4f5276414d Merge pull request #364 from reunanen/set-cuda-propagate-host-flags-on 
Fix Visual Studio build by not setting CUDA_PROPAGATE_HOST_FLAGS to OFF
 2016-12-19 11:43:16 -05:00
Juha Reunanen 2d23e4ddba Problem: Visual Studio linker errors resulting from mixed /MT and /MD settings 
Solution: don't set CUDA_PROPAGATE_HOST_FLAGS to OFF when using Visual Studio
 2016-12-19 17:07:43 +02:00
Juha Reunanen 33e5889399 Workaround for #363: do not set CUDA_HOST_COMPILER for the CUDA test 2016-12-19 16:10:38 +02:00
Davis King bf36a846de Changed build order to see if it prevents travis from crashing. 2016-12-18 23:35:25 -05:00
Davis King 093f709a9a Minor improvement to example 2016-12-18 20:52:45 -05:00
Davis King ab1575da20 more cleanup 2016-12-18 13:20:37 -05:00
Davis King b87ecad51e Improved example 2016-12-18 13:10:13 -05:00
Davis King fd13230486 Finished the more complex metric learning example and added some example data. 2016-12-17 20:41:36 -05:00