use static forward and backward methods

NVIDIA / flownet2-pytorch

Pytorch implementation of FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

Other

3.12k stars 739 forks source link

use static forward and backward methods #207

Closed bigrobinson closed 4 years ago

bigrobinson commented 4 years ago

Use static methods for autograd extension to relieve deprecation warning in Pytorch 1.4.

fitsumreda commented 4 years ago

thanks!

bigrobinson commented 4 years ago

Not at all. A big thanks to you guys!

On Wed, Apr 1, 2020 at 5:05 PM Fitsum Reda notifications@github.com wrote:

thanks!

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/NVIDIA/flownet2-pytorch/pull/207#issuecomment-607513018, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKCJQUCEOIDNMO3OESWFIT3RKO3B5ANCNFSM4LY5V4IQ .

suyashhchougule commented 3 years ago

Use static methods for autograd extension to relieve deprecation warning in Pytorch 1.4.

Hi,I am struck with this error. But , I dont quite understand what has to be done . Could you elaborate more ?

AlgorithmicPlanet commented 1 year ago

RuntimeError: Legacy autograd function with non-static forward method is deprecated. Please use new-style autograd function with static forward method.