DL-Art-School/codes
James Betker 4328c2f713 Change default ReLU slope to .2 BREAKS COMPATIBILITY
This conforms my ConvGnLelu implementation with the generally accepted negative_slope=.2. I have no idea where I got .1. This will break backwards compatibility with some older models but will likely improve their performance when freshly trained. I did some auditing to find what these models might be, and I am not actively using any of them, so probably OK.
2020-12-19 08:28:03 -07:00
..
.idea IDEA update 2020-05-19 09:35:26 -06:00
data More refactor changes 2020-12-18 09:24:31 -07:00
models Change default ReLU slope to .2 BREAKS COMPATIBILITY 2020-12-19 08:28:03 -07:00
scripts GLEAN! 2020-12-18 16:04:19 -07:00
trainer Add trainer readme 2020-12-18 16:52:16 -07:00
utils Update other docs with dumb config options 2020-12-18 16:21:28 -07:00
multi_modal_train.py More adjustments to support distributed training with teco & on multi_modal_train 2020-10-27 20:58:03 -06:00
process_video.py More refactoring 2020-12-18 09:18:34 -07:00
requirements.txt Update requirements, add image_patch_classifier tool 2020-12-16 09:42:50 -07:00
test_image_patch_classifier.py More refactoring 2020-12-18 09:18:34 -07:00
test.py More refactoring 2020-12-18 09:18:34 -07:00
train.py Update other docs with dumb config options 2020-12-18 16:21:28 -07:00