README.md

metadata

language: en
license: mit
library_name: pytorch

Knowledge Continuity Regularized Network

Trainer Hyperparameters:

• lr = 1e-05
• per_device_batch_size = 8
• gradient_accumulation_steps = 2
• weight_decay = 1e-09
• seed = 42

Regularization Hyperparameters

• numerical stability denominator constant = 0.01
• lambda = 0.001
• alpha = 2.0
• beta = 1.0

Extended Logs:

eval_loss	eval_accuracy	epoch
5.407	0.792	1.0
5.131	0.792	2.0
4.966	0.792	3.0
4.796	0.792	4.0
4.710	0.792	5.0
4.653	0.792	6.0
4.534	0.792	7.0