Skip to content
This repository has been archived by the owner on Dec 29, 2022. It is now read-only.

train detail #27

Open
Xue9901 opened this issue Apr 7, 2020 · 0 comments
Open

train detail #27

Xue9901 opened this issue Apr 7, 2020 · 0 comments

Comments

@Xue9901
Copy link

Xue9901 commented Apr 7, 2020

  Hello!thank you for the wonderfull work!  I'm very interested in your project ! I'm training this model for about one day and one night on GPU, but the loss seems to be still sometimes very big and sometimes very small ,and how long did you take for the loss to converge during the training?  
  ps. i  set the batchsize to 8,with learning rate= 0.0002,β1 = 0.9,β2 = 0.999
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant