search

Bibyutatsu / proGENTRL

Pytorch Lightning implementation of GENTRL
https://github.com/Bibyutatsu/GENTRL
MIT License
3 stars 1 forks source link

Is there any solution on CUDA out of memory errors without Upgrade my GPU system (( buy new GPU with higher memory minimum 8 Gb) #5

Open Dharmogata opened 4 years ago

Dharmogata commented 4 years ago

Hello,Bibyutatsu

I am writing you because I am stuck in implemention of the code in github.

could you please let me know how to solve this CUDA out of memory errors without Upgrade my GPU system ( buy new GPU with higher memory minimum 8 Gb)?

스크린샷, 2020-09-24 10-07-31 스크린샷, 2020-09-24 10-27-37 스크린샷, 2020-09-24 10-18-17

Thank you for this great work!

Bibyutatsu commented 4 years ago

Hi @gwanseum , You can use precision 16 or a lower batch size.

Please refer here for precision related problems.

You can also use Colab or Kaggle for training as they give you a 8 GB free GPU.