search

StefOe / indrnn-pytorch

pytorch implementation of Independently Recurrent Neural Networks https://arxiv.org/abs/1803.04831
121 stars 33 forks source link

In IndRNNCell, what's the reason for doing the 'gradient_clip'? #6

Open MLjian opened 6 years ago

MLjian commented 6 years ago

I think the check_bounds has done the same operation.

StefOe commented 5 years ago

"Gradient clip" clips the gradient, while the actual weight is altered by check_bounds.