New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

为什么显存占用这么高Why is the memory usage so high #538

Open

BucolicLife opened this issue Nov 1, 2023 · 0 comments

BucolicLife commented Nov 1, 2023

INFO:NanoDet:Train|Epoch1/100|Iter200(201/3000)| mem:1.4G| lr:4.00e-04| loss_qfl:0.3616| loss_bbox:1.0301| loss_dfl:0.4526| aux_loss_qfl:0.4193| aux_loss_bbox:0.5471| aux_loss_dfl:0.2657|
[NanoDet][11-01 10:05:29]INFO:Train|Epoch1/100|Iter250(251/3000)| mem:15.3G| lr:5.00e-04| loss_qfl:0.8047| loss_bbox:0.8445| loss_dfl:0.3974| aux_loss_qfl:0.6471| aux_loss_bbox:0.7851| aux_loss_dfl:0.3845|
INFO:NanoDet:Train|Epoch1/100|Iter250(251/3000)| mem:15.3G| lr:5.00e-04| loss_qfl:0.8047| loss_bbox:0.8445| loss_dfl:0.3974| aux_loss_qfl:0.6471| aux_loss_bbox:0.7851| aux_loss_dfl:0.3845|
同一个epoch显存从1.4G猛涨到15.3G
训练集3000条样本，验证集300条，图像大小[480,640]
In the same epoch, the video memory jumped from 1.4G to 15.3G.
There are 3000 samples in the training set, 300 samples in the verification set, and the image size is [480, 640]

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment