You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for this work. I am trying to train RCAN from scratch on by dataset. I've noticed that my training time increases when I increase the number of GPU. For example when I used 4 GPU and 36 batchs I had a training time of 70s for 3200 images.
But when I used only 1 GPU for the same number of batchs, I had a training time of 31s for 3200 images.
I really can't explain why.
Do you have an idea ?
Thank you a lot
Rehan JHUBOO
The text was updated successfully, but these errors were encountered:
Hello,
Thank you for this work. I am trying to train RCAN from scratch on by dataset. I've noticed that my training time increases when I increase the number of GPU. For example when I used 4 GPU and 36 batchs I had a training time of 70s for 3200 images.
But when I used only 1 GPU for the same number of batchs, I had a training time of 31s for 3200 images.
I really can't explain why.
Do you have an idea ?
Thank you a lot
Rehan JHUBOO
The text was updated successfully, but these errors were encountered: