6c43f6ba38
Training: Pass imgDim to batchRepresent. Add lfwDir option.
...
For #112 .
2016-03-19 15:44:47 -04:00
b30d6d9461
Training Plotting code: Add minor gridlines.
2016-03-18 14:10:53 -04:00
bba911c802
Replace satitize by clearState()
2016-03-16 18:20:11 +01:00
e7c574829d
Merge pull request #108 from melgor/cudnn_convert
...
Replace nn_to_cudnn by cudnn.convert #106
2016-03-16 11:54:20 -04:00
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
f63acb637e
Replace nn_to_cudnn by cudnn.convert #106
2016-03-16 15:38:06 +01:00
6fd4ac8604
Training: Sample from all of a person's images.
...
This fixes a bug in the training code that caused
only the first (same) `nSamplesPerClass` images in each
class to be sampled every time rather than sampling
from all of the person's images.
2016-03-15 16:27:09 -04:00
e2d7513ff5
Replace nn_to_cudnn by cudnn.convert #106
2016-03-11 11:09:23 +01:00
71b6bbf765
Fix flake warning.
2016-03-08 20:39:03 -05:00
665cb4c9d7
Training plot-loss.py: Improve triplet loss plot bounds.
2016-03-08 18:36:58 -05:00
04d23481fb
Fix typo.
2016-03-06 21:15:10 -05:00
4616f61ff8
DNN Training: Halve default epochSize.
2016-03-06 20:03:54 -05:00
f17a36c2fd
plot-loss: Import sys.
2016-03-06 19:56:12 -05:00
f112398178
DNN Training: Plot LFW accuracies for #100 .
2016-03-06 19:46:33 -05:00
bbff0a01dd
Fix tests for training the DNN.
2016-03-06 19:45:37 -05:00
f5766412d2
NN Training: Add LFW validation for #100 .
2016-03-04 18:30:39 -05:00
e4c4aef3e4
Training: Zero pad experiment work directory numbers.
2016-03-04 18:09:55 -05:00
540b7d03c6
Fix luacheck warning.
2016-02-28 15:08:50 -05:00
49d195355b
Training: Remove test code from data loader.
2016-02-28 13:48:22 -05:00
e987b15b22
Move old test code to attic for #99 .
2016-02-27 17:50:11 -05:00
9b04bca65d
train.lua: Pass on comments.
2016-01-16 23:50:54 -05:00
51b01c200c
Train: Fix model saving logic for out-of-memory errors.
...
Thanks @kleinsound
2016-01-16 16:37:35 -05:00
9d1bd99579
Update copyright notices.
...
sed -i "/[Cc]opyright.*2015.*Carnegie Mellon University/s/2015/2015-2016/;" $(git grep '[Cc]opyright.*Carnegie Mellon University' | sed 's/:.*//' | uniq)
2016-01-13 14:52:52 -05:00
8bbfd4da05
Training: Skip battch if nTripsFound == 0.
2016-01-13 14:32:43 -05:00
d9c602c449
Add cuda conditionals to test.lua.
2016-01-12 17:43:50 -05:00
e5277e916c
sanitize: Remove unused imports.
2016-01-12 17:40:01 -05:00
dd5d0d11bd
Training: Remove unused 'fbnn' import.
2016-01-12 17:30:06 -05:00
6409661f5f
Remove unused fbnn dependency.
2016-01-12 14:32:54 -05:00
0479d0e345
Finish tensor size checks for #36 .
2016-01-11 16:44:50 -05:00
d0b40686a2
Training: OpenFaceOptim: Correct Tensor type.
2016-01-11 16:44:32 -05:00
5f78673b7d
Move imgDim to models.
2016-01-11 16:38:24 -05:00
c5b0accd4d
Training: Add conditionals for cuda portions.
2016-01-11 16:17:36 -05:00
15cc5fef00
Training: Add CPU execution for tests.
2016-01-11 16:17:36 -05:00
ff06eff893
Docker: Move Python dependencies to Ubuntu package binaries.
2016-01-11 16:12:45 -05:00
6a5d95d451
Fix luacheck warning in sanitize.
2016-01-11 07:48:01 -05:00
8155840989
Training: Use e-lab's torch-toolbox sanitize.
2016-01-09 17:17:30 -05:00
697fac07c1
Training: Output the number of model parameters.
2016-01-08 11:18:44 -05:00
22cf225923
Training: By default use an epoch size of 500.
2016-01-08 11:00:17 -05:00
6246fada83
Update code to nn4.v2 for #55 .
2016-01-07 18:28:11 -05:00
4821cb6ee2
Fix flake8 warnings.
2016-01-06 17:19:37 -05:00
2b8f03c908
Update lfw-latest.sh.
2016-01-06 17:02:30 -05:00
2769411d56
Add @melgor's training improvements for #55
2016-01-06 17:02:30 -05:00
4098a586ae
Training: Add script to run lfw evaluation on the latest model.
2015-12-28 16:54:35 -05:00
ff51b2cbde
Fix luacheck warnings.
2015-12-27 09:09:14 -05:00
57118ac28f
Fix luacheck warnings.
2015-12-27 08:41:49 -05:00
81a62c6e13
Fix broken links.
2015-12-26 17:11:44 -05:00
dceb6a910b
flake8 improvements.
2015-12-24 11:45:00 -05:00
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
eea992b458
Mention Python dependencies for training.
2015-12-17 11:43:00 -05:00
bf38abe8f9
Remove unused cudnn dependency.
2015-12-17 09:04:46 -05:00
Brandon Amos