microsoft
/
caffe
зеркало из https://github.com/microsoft/caffe.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
284 коммитов 3 Ветки 11 Теги 51 MiB
Граф коммитов

284 Коммитов

Автор SHA1 Сообщение Дата
Yangqing Jia 3b25abce88 Merge branch 'master' of github.com:Yangqing/caffe 2013-11-19 14:03:56 -08:00
Yangqing Jia d221841c0a makefile: fixed the nvcc path 2013-11-19 13:39:20 -08:00
Yangqing Jia 391cccfa6e io.cpp: changed the imread flag to cv::IMREAD_COLOR 2013-11-19 13:37:28 -08:00
Yangqing Jia db5fb24356 Merge branch 'master' of github.com:Yangqing/caffe 2013-11-19 11:06:12 -08:00
Yangqing Jia 46a431ac74 added explanation notes to syncedmem 2013-11-19 09:08:31 -08:00
Yangqing Jia e08801e728 imagenet_solver.prototxt: added back the missing test line, not sure when it slipped off the repo 2013-11-19 08:51:43 -08:00
Yangqing Jia 4dae5e8fb7 Merge branch 'master' of github.com:Yangqing/caffe 2013-11-18 16:06:37 -08:00
Yangqing Jia 91254090bd imagenet deploy prototxt: instead of having a data layer, this network proto takes an external blob called data. The shape is hardcoded to accompany the imagenet python wrapper at python/caffe/imagenet/wrapper.py. You need to change the shape if you intend to use other shapes. 2013-11-18 16:06:33 -08:00
Jeff Donahue 3b187e0f6c print status message on error opening leveldb 2013-11-17 14:39:42 -08:00
Jeff Donahue 734a3fda0d Merge branch 'master', remote-tracking branch 'origin' 2013-11-17 13:45:22 -08:00
Yangqing Jia 01686065c6 removed the StillFresh function that was used to make sure training does not happen with an earlier version of code. 2013-11-16 18:41:28 -08:00
Yangqing Jia e24b1dd22b Made a major change: when initializing a network, the input size are no longer provided by an additional vector of blobs, but should be specified in the netparameters proto by the field "input_dim". This avoid the often awkward code of creating a dummy input vector just for the sake of initializing the network. 2013-11-15 16:58:53 -08:00
Yangqing Jia f54e9779ef wrapper update 2013-11-15 08:02:48 -08:00
Yangqing Jia 6def792e66 finetune code 2013-11-14 21:12:23 -08:00
Yangqing Jia fa7b43062b pycaffe: added a temporary numpy 1.6 compile solution 2013-11-14 14:51:27 -08:00
Yangqing Jia a5f2611988 misc update 2013-11-14 14:33:24 -08:00
Yangqing Jia ff9c69c77c pycaffe update and imagenet wrapper 2013-11-13 16:31:40 -08:00
Yangqing Jia b4fa660d5b pycaffe update 2013-11-13 11:29:26 -08:00
Yangqing Jia 26f5a140e1 linecount improvement 2013-11-12 14:31:09 -08:00
Yangqing Jia 78cca4b90d ilsvrc 2012 mean 2013-11-12 14:21:24 -08:00
Yangqing Jia 90efd7b968 removed obsolete scripts 2013-11-12 14:16:34 -08:00
Yangqing Jia eca3e9e774 python reorganization 2013-11-12 14:15:33 -08:00
Yangqing Jia a8785707c3 changed makefile, and removed the no longer needed cuda convnet translator. 2013-11-12 14:14:02 -08:00
Yangqing Jia 6e26269e9d changed the python file paths 2013-11-12 14:05:52 -08:00
Yangqing Jia 32b7dded14 added pycaffe wrapper. Preparing to clean the structure 2013-11-12 13:49:37 -08:00
Yangqing Jia 652d744360 Merge branch 'master' of github.com:Yangqing/caffe 
Conflicts:
	Makefile
 2013-11-11 17:08:27 -08:00
Yangqing Jia 76bf486e3d remove python deps 2013-11-11 17:07:32 -08:00
Yangqing Jia c8e7cce731 stochastic pooling: avoid nan 2013-11-11 17:03:27 -08:00
Yangqing Jia 9563a78b21 io.hpp bugfix 2013-11-11 17:02:38 -08:00
Yangqing Jia 9f0bfdd036 added infogain loss layer 2013-11-11 11:40:31 -08:00
Yangqing Jia a3875f5eb2 bugfix 2013-11-11 11:10:39 -08:00
Yangqing Jia 22d82ffe15 misc update 2013-11-08 13:50:12 -08:00
Yangqing Jia 8e9916ad0d Merge branch 'master' of github.com:Yangqing/caffeine 2013-11-07 11:29:22 -08:00
Yangqing Jia ade9664cba ilsvrc 2012 train file list 2013-11-07 11:29:18 -08:00
Yangqing Jia a48147c1b8 distributed solver still having bugs. Pausing for now... 2013-11-07 11:06:20 -08:00
Yangqing Jia a3eb62af9b distributed solver: small fix, still bugs 2013-11-06 16:38:34 -08:00
Yangqing Jia 591c36bba0 distributed server update. bug in synchronous connections. 2013-11-06 15:27:26 -08:00
Yangqing Jia 5d14903df5 data layer: random skip 2013-11-06 14:02:33 -08:00
Yangqing Jia 185385b87f working asynchronous sgd code. may have errors. 2013-11-06 11:28:17 -08:00
Yangqing Jia 64e28ba82b distributed solver - not checked thoroughly 2013-11-04 11:14:36 -08:00
Yangqing Jia f19ce20bc6 solver minor change 2013-11-04 11:13:19 -08:00
Yangqing Jia e334ea9f55 accuracy layer: also produce logprob 2013-11-01 10:23:07 -07:00
Yangqing Jia fb9968f0b5 imagenet solver 2nd try 2013-10-31 17:02:22 -07:00
Yangqing Jia 19a03e1136 imagenet test prototxt 2013-10-31 16:59:44 -07:00
Yangqing Jia 82b912be84 solver restructuring: now all prototxt are specified in the solver protocol buffer 2013-10-31 16:52:22 -07:00
Yangqing Jia 25a865cd8b stochastic pooling test 2013-10-31 12:18:35 -07:00
Yangqing Jia b9966b84f8 test convolution more thoroughly 2013-10-31 11:10:54 -07:00
Yangqing Jia b9d908bf2f script to convert cifar data 2013-10-30 16:52:34 -07:00
Yangqing Jia 236eef6755 script to get cifar 2013-10-30 16:47:58 -07:00
Yangqing Jia cfd6e1dbf2 test net scripts 2013-10-30 15:55:10 -07:00