There is no train.py
yangpanquan opened this issue · comments
空山夜雨 commented
Hi, it's a good work in the real-time semantic segmentation.But, there is no training file here. could you upload it in your free time? Besides, there is another question. what does the meaning of "The training settings require 8 GPU with at least 11GB memory."?
Qi Song commented
Sorry for the uploading mistake. The train.py is now uploaded. To train AttaNet_resnet18 or AttaNet_light with 8 images per GPU, 11GB memory is needed.
空山夜雨 commented
Thank you! Moreover, I found there is another question in the cityscapes.py . There is no HorizontalFlip function,Rotation function,and RandomScale function.Can you upload them?