Commit Graph

206 Commits

Author SHA1 Message Date
Alexander Alekhin
73a8e65cbd Merge pull request #5285 from ilya-lavrenov:ml5 2015-09-02 11:39:07 +00:00
Alexander Alekhin
ae0d428b62 Merge pull request #5281 from ilya-lavrenov:ml2 2015-09-02 11:38:45 +00:00
Vadim Pisarevsky
5603617b7c Merge pull request #5284 from ilya-lavrenov:ml4 2015-09-01 12:31:46 +00:00
Vadim Pisarevsky
43aae3fee8 Merge pull request #5283 from ilya-lavrenov:ml3 2015-09-01 12:31:05 +00:00
Ilya Lavrenov
1b8c2589c0 fixed memory leak in GBTrees 2015-09-01 14:38:11 +03:00
Ilya Lavrenov
887736bcd4 fixed "Conditional jump or move depends on uninitialised value(s)" in GBD 2015-09-01 13:22:49 +03:00
Ilya Lavrenov
3a1bb93340 release filestorage before exception 2015-09-01 13:04:33 +03:00
Ilya Lavrenov
dfb49097e3 fixed memory leak in ANN 2015-09-01 12:29:52 +03:00
Ilya Lavrenov
d7bb1025f3 fixed memory leak in ml module 2015-09-01 11:26:25 +03:00
Maksim Shabunin
447b8bf58a Fixing possible corruption for big training data sizes 2015-08-10 12:36:27 +03:00
Maksim Shabunin
e23ba9215d Fixed memory leak in ML module
"int * CvMLData::sample_idx" was not freed
2014-12-30 15:20:24 +03:00
Maksim Shabunin
91e21847e2 Some changes to make migration to 3.0 easier 2014-12-16 13:00:31 +03:00
Benoit Seguin
084835ec30 Fixing a line where integer data was treated as floating point data. 2014-09-18 14:03:13 +02:00
unknown
ade46bd428 Fixed typos in comments 2014-06-26 16:29:45 +02:00
berak
2bacd8b702 2 fixed unassigned reshapes in em (#3674) 2014-05-28 10:37:16 +02:00
Vladislav Vinogradov
7e2f7f45d7 fix bug #3690
removed invalid condition, it is always false
2014-05-14 10:47:28 +04:00
Andrey Pavlenko
32bc89f094 Merge pull request #2107 from nghiaho12:nn_doc 2014-01-21 19:11:12 +04:00
Daniil Osokin
5d2edced26 Added throwing exception when saving untrained SVM model 2014-01-13 13:50:30 +04:00
Ilya Lavrenov
6b9ebcbf3d deleted extra semicolons 2014-01-07 02:52:30 +04:00
Nghia Ho
d3e24f3cbf Improved documentation for neural network 2014-01-03 14:18:07 +11:00
Roman Donchenko
06f6f74e61 Merge pull request #1839 from pmiecielica:memory_leaks 2013-11-25 12:17:08 +04:00
Piotr Miecielica
923698d700 Fixes of some memory leaks 2013-11-21 18:55:09 +01:00
Roman Donchenko
cf00349b5b Gave CvSVM a private copy constructor & assignment operator.
The implicitly generated ones don't work properly, and I don't want to
write proper ones. :-)

"Fixes" <http://code.opencv.org/issues/3358>.
2013-11-20 14:59:42 +04:00
Alexander Alekhin
f63b0dfaf4 svm: disable failed tests 2013-10-07 18:17:10 +04:00
Csaba Kertesz
6fb3b4f41c Fix a memory leak in CvNormalBayesClassifier::train() 2013-09-20 22:36:49 +02:00
Roman Donchenko
e9a28f66ee Normalized file endings. 2013-08-21 18:59:25 +04:00
Roman Donchenko
f55740da70 Deleted all trailing whitespace. 2013-08-21 18:59:24 +04:00
Roman Donchenko
d9f71b762f Deleted almost all of the precomp.cpp files.
Looks like we don't actually use them; when creating precompiled headers
with Visual C++, a dummy .cpp is created at build time.
2013-08-08 12:43:07 +04:00
StevenPuttemans
ed76b2f98f Like mentioned by Andrei Pavlenko after merging pullrequest #1206, a wrong
ReST directive was used. Also fixed some other ReST directives that were
not correct and removed some warnings during buildbot checks.
2013-08-07 14:49:33 +02:00
StevenPuttemans
f28496c628 Added sample links in documentation for python2, ocl and gpu 2013-08-02 14:05:08 +02:00
Roman Donchenko
ac39bfb4cc Remove HAVE_CVCONFIG_H - it's always defined. 2013-07-10 16:02:07 +04:00
Vladislav Vinogradov
e433145b7e fix for Bug #3085:
weights array is only allocated for (l_count+1) elements,
but then weights[l_count+1] element is accessed.
2013-06-13 10:22:56 +04:00
Roman Donchenko
29b13ec1de Replaced most of the usages of parallel_for with that of parallel_for_.
This should allow many algorithms to take advantage of more parallelization
technologies.
2013-05-31 12:22:04 +04:00
Roman Donchenko
4143071e22 In CvBoost, delegated update_weights's implementation to a helper method.
This allows subclasses of CvBoost to override initial weights that
update_weights uses without duplicating its entire implementation.
2013-04-11 13:50:10 +04:00
Vadim Pisarevsky
e01335bf47 fixed buffer size and restored the use of compressed files in ml's load_save tests. 2013-03-22 01:53:41 +04:00
Vadim Pisarevsky
4331f76d18 add hack to disable optimization of linear svms; improved precision of optimize_linear_svm; add the relevant test, which however requires some big database (so it's disabled by default) 2013-03-21 19:17:59 +04:00
Vadim Pisarevsky
63a5587d0d exploring possible bug in optimize_linear_svm 2013-03-21 17:00:08 +04:00
Vadim Pisarevsky
a06af5ca25 dramatic speedup of SVM::predict in the case of linear SVM 2013-03-19 18:28:17 +04:00
Andrey Kamaev
5691fd3abe Merge pull request #503 from taka-no-me:fix_docs 2013-02-19 18:01:51 +04:00
Andrey Kamaev
ff7b604a25 Fix several inconsistent declarations identified by check_doc2 script 2013-02-19 16:22:20 +04:00
Niels Gerlif Myrtue
7697d79d93 Added the predict_all function wrapper for python that was missing from the documentation
Copied the docstring description
2013-02-19 13:11:30 +11:00
Andrey Kamaev
9c939a8dcf Merge pull request #420 from asmaloney:check-mem-alloc 2013-02-04 12:56:36 +04:00
Andrey Kamaev
e2536f1c35 Merge pull request #395 from LeonidBeynenson:fix_ml_large_data_bug__2.4 2013-02-04 11:54:43 +04:00
Andy Maloney
b497380a68 Check memory allocation
Initialize local variables
2013-02-02 08:33:40 -05:00
LeonidBeynenson
87b0126e0d Fixed dummy warning. 2013-02-01 16:16:43 +04:00
LeonidBeynenson
013d54c230 Changed types of some variables from int64 back to int.
Also corrected some indexes to be size_t.
2013-02-01 14:41:14 +04:00
LeonidBeynenson
6de422701a Made changes to allow ml module to work with big data. 2013-01-31 19:37:20 +04:00
Daniil Osokin
fe5b9df45f Fixed multiple issues in docs (bug #2410) 2013-01-31 17:34:40 +04:00
Alexander Smorkalov
8cb0343f4c Documentation build warning fixes.
Invalid links and refences fixed.
SpaeseMat class documentation updated.
2013-01-14 17:58:27 +04:00
Maria Dimashova
0e2958e888 fixed CvBoost::predict
(Sometimes sample_data became bad because the buffer with its values was deallocated early).
2012-12-29 00:30:52 +04:00