WXinlong / SOLO

SOLO and SOLOv2 for instance segmentation, ECCV 2020 & NeurIPS 2020.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Memory usage problem when testing

pmz1997 opened this issue · comments

commented

Thank you for your amazing work.
I trained and tested light-weight model on my own dataset. However, I found GPU memory usage in testing is much higher than that in training(2.7GB with bs 2 vs 8.8GB with bs 1). I trained model with single GPU and tested it using test_ins.py
Could you kindly help me to solve this issue?