Commit Graph

18 Commits

Author SHA1 Message Date
melgor 0995e86b7e Added Multi-GPU support #106 2016-07-04 22:23:37 +02:00
Brandon Amos 2c6ddfba2d Training: Model: Use opt.imgDim. 2016-06-16 20:12:21 -04:00
Brandon Amos fea2fb41d8 Resolve nan issues for #127 (hopefully) 2016-06-13 13:50:38 -04:00
Brandon Amos 21f1bcc5c4 Temporarily remove optimize_net for #127. 2016-06-07 18:14:13 -04:00
Brandon Amos 66efcb706b Model: Use opt.imgDim for retraining
\cc @melgor

Ideally I should have made imgDim part of the model.
This works for now, but users training will need to be
careful about this size.
2016-04-05 18:02:45 -04:00
melgor bf6a466484 Added optnet for reducing memory consumption 2016-03-30 23:21:13 +02:00
melgor 7d21caf004 Merge branch 'cudnn_convert' of https://github.com/melgor/openface into cudnn_convert
Conflicts:
	training/model.lua
	training/train.lua
2016-03-16 16:15:05 +01:00
melgor f63acb637e Replace nn_to_cudnn by cudnn.convert #106 2016-03-16 15:38:06 +01:00
melgor e2d7513ff5 Replace nn_to_cudnn by cudnn.convert #106 2016-03-11 11:09:23 +01:00
Brandon Amos 6409661f5f Remove unused fbnn dependency. 2016-01-12 14:32:54 -05:00
Brandon Amos 0479d0e345 Finish tensor size checks for #36. 2016-01-11 16:44:50 -05:00
Brandon Amos 15cc5fef00 Training: Add CPU execution for tests. 2016-01-11 16:17:36 -05:00
Brandon Amos 697fac07c1 Training: Output the number of model parameters. 2016-01-08 11:18:44 -05:00
Brandon Amos 6246fada83 Update code to nn4.v2 for #55. 2016-01-07 18:28:11 -05:00
Brandon Amos 2769411d56 Add @melgor's training improvements for #55 2016-01-06 17:02:30 -05:00
Brandon Amos e90694fcba nn4: Add LRN and remove 5x5 convolutions for #55.
Reduces the number of parameters from 7472144 to 6959088.
2015-12-20 22:29:55 -05:00
Brandon Amos bf38abe8f9 Remove unused cudnn dependency. 2015-12-17 09:04:46 -05:00
Brandon Amos d8790408d1 Initial commit of training code from development repo. 2015-09-23 19:49:45 -04:00