search

xuanyuzhou98 / SqueezeSegV2

Implementation of SqueezeSegV2, Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloud
BSD 2-Clause "Simplified" License
233 stars 69 forks source link

Learning rate changes abnormally #3

Open MoonWolf9067 opened 5 years ago

MoonWolf9067 commented 5 years ago

Hi! Thank you for sharing your work. When I look at the training effect in the tensorboard, the learning rate shows the following trend. The learning rate change is different from the exponential decay trend set in the code. Can you give me some suggestions? 3333

xuanyuzhou98 commented 5 years ago

@MoonWolf9067 I tried in a new environment and the learning rate follows exponential decay correctly. Could you repull the code and provide more information about your changes?

MoonWolf9067 commented 5 years ago

Thank you for your reply, I have understood this question. The learning rate indeed follows exponential decay correctly.