AntoineH/smart_augmentation: 'Smart_aug' objective is to make data augmentation differentiable, thus allowing to learn its parameters, with 'Higher' objects, such as 'Data_aug' classes. The meta-learning of the data augmentation parameter is performed jointly with the training of the model. Thus it minimize the overhead compared to other data augmentation learning techniques.

mirror of https://github.com/AntoineHX/smart_augmentation.git synced 2025-05-04 12:10:45 +02:00

'Smart_aug' objective is to make data augmentation differentiable, thus allowing to learn its parameters, with 'Higher' objects, such as 'Data_aug' classes. The meta-learning of the data augmentation parameter is performed jointly with the training of the model. Thus it minimize the overhead compared to other data augmentation learning techniques.

Find a file

Harle, Antoine (Contracteur) 994d657a28 Dataugv5- Modification des TF pour propagation du gradient (mag)		2019-11-18 12:53:23 -05:00
FAR-HO	Initial Commit	2019-11-08 11:28:06 -05:00
Gradient-Descent-The-Ultimate-Optimizer	Initial Commit	2019-11-08 11:28:06 -05:00
higher	Dataugv5- Modification des TF pour propagation du gradient (mag)	2019-11-18 12:53:23 -05:00
PBA	Premieres modifications pour LeNet sur PBA	2019-11-14 17:59:53 -05:00
.gitignore	Ignore sample dir	2019-11-11 17:03:31 -05:00