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
I have bought colab pro to run image to 3D. Things went well until I ran in training stage. I follow README.md to run the command using zero123 backend , but when it trained about 2 minutes , system RAM is not enough, and GPU RAM seems not being used as picture showed below, how can I solve the problem?
You can change your execution type to A100 and then you will have 83.5 GB system RAM and 40 GB GPU RAM. In this execution type, Image to 3D works really well !
I have bought colab pro to run image to 3D. Things went well until I ran in training stage. I follow README.md to run the command using zero123 backend , but when it trained about 2 minutes , system RAM is not enough, and GPU RAM seems not being used as picture showed below, how can I solve the problem?
![螢幕擷取畫面 2023-07-03 212827](https://private-user-images.githubusercontent.com/106028807/250572812-3581421f-8726-4432-bd60-b64b7defc11e.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MTg1NTc5NTAsIm5iZiI6MTcxODU1NzY1MCwicGF0aCI6Ii8xMDYwMjg4MDcvMjUwNTcyODEyLTM1ODE0MjFmLTg3MjYtNDQzMi1iZDYwLWI2NGI3ZGVmYzExZS5wbmc_WC1BbXotQWxnb3JpdGhtPUFXUzQtSE1BQy1TSEEyNTYmWC1BbXotQ3JlZGVudGlhbD1BS0lBVkNPRFlMU0E1M1BRSzRaQSUyRjIwMjQwNjE2JTJGdXMtZWFzdC0xJTJGczMlMkZhd3M0X3JlcXVlc3QmWC1BbXotRGF0ZT0yMDI0MDYxNlQxNzA3MzBaJlgtQW16LUV4cGlyZXM9MzAwJlgtQW16LVNpZ25hdHVyZT0zNWQzMmY2MDc3NDdiMmM3Y2IxOWU2N2I5ZDdlMTllYmNmMzE1OGNhZDk5MmEwMzgzNzlkY2ViMWMyMDhjZDhiJlgtQW16LVNpZ25lZEhlYWRlcnM9aG9zdCZhY3Rvcl9pZD0wJmtleV9pZD0wJnJlcG9faWQ9MCJ9.ckupC-DvCuF1iDb7CVedKDCgc9l9I8LcOPfAOyOn9d0)
Code
The code I ran :
https://colab.research.google.com/drive/1jP_eM8FMeyygv0NfmQsEcThjmy91sTvw?authuser=1#scrollTo=RH-8W-VT_XiU
Environment
google colab pro
The text was updated successfully, but these errors were encountered: