During training, use dilate_loss for early stopping

[pratham16cse]

• When base model is DILATE, compute dilate_loss(..) on val data during training.
• A separate metric for early stopping for each model based on sutaibility.

[pratham16cse]

• Added dilate_loss(..) for early stopping of DILATE base model.
• For all other models, MSE is used for early stopping.