Chung-Yi

I want to ask that my least training loss is between 0.8-1.5. Does anyone know how to decrease the loss. Although it has nice performance at most condition.

affinex = tf.stack([tf.maximum(affine_pred[...,0],0.),affine_pred[...,1],affine_pred[...,2]],3) affiney = tf.stack([affine_pred[...,3],tf.maximum(affine_pred[...,4],0.),affine_pred[...,5]],3) affinex and affiney mismatch with the paper 

val_loss can not decrease

2

I faced some problem when I trained. At the beginning, both the loss and val_loss decreased step by step, but then only loss decreased val_loss frozen between 24 and 25....