Giters

zhanggang001 / RefineMask

RefineMask: Towards High-Quality Instance Segmentation with Fine-Grained Features (CVPR 2021)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

About the samples_per_gpu and learning rate of Cityscapes dataset

AGanPoi opened this issue · comments

commented

I see the samples_per_gpu of Cityscapes is change to 2, but the learning rate is still set to 0.01. I understand the modification of samples_per_gpu of COCO dataset, for the batchsize is 16 (8 GPU * 2 samples_per_gpu) and corresponding to learning rate of 0.02. But I wonder when the samples_per_gpu of Cityscapes dataset is 2, the learning rate should also change?

commented

The total batch size for training Cityscapes is 8 by default. So if the samples_per_gpu is set to 2, the number of GPUs should be set to 4.

commented