Delay-Xili / F-Clip

The official PyTorch implementation of the paper: *Xili Dai, Xiaojun Yuan, Haigang Gong, Yi Ma. "Fully Convolutional Line Parsing." *.
MIT License
150 stars 21 forks source link

about training time #7

Closed chenyuyuyu closed 1 year ago

chenyuyuyu commented 3 years ago

Hello, I want to know how long it took you to train FClip-HR? It took me more than seven days. Is there any way to reduce the training time?

Delay-Xili commented 3 years ago

Hi, If you are not pursuing the state of the art performance, just set the training epoch to 35 and make the learning rate decay at 25 and 30 respectively, that will be enough for you to obtain acceptabel results.

lucy3589 commented 1 year ago

Hi, If you are not pursuing the state of the art performance, just set the training epoch to 35 and make the learning rate decay at 25 and 30 respectively, that will be enough for you to obtain acceptabel results.

if train HG1_D2 model. How many epochs need to be trained?