DeepOcclusion icon indicating copy to clipboard operation
DeepOcclusion copied to clipboard

Published 20 hours ago • verified publisher icon pierrebaque

About background training

Open lc-git opened this issue 6 years ago • 1 comments

Hi,

I have two questions about background model training:

  1. I wonder whether I need to train from scratch or train from pretrained model? The pretrained model will make the convergence faster?
  2. approximately how many images do we need to train my own model?

lc-git avatar Mar 21 '18 13:03 lc-git

Hi, It seems that you have done some training on your own dataset based on this project, so have you got some progresses recently? I had some troubles when trying to retrain the models based on my own dataset. Thanks

soullessrobot avatar Feb 22 '19 14:02 soullessrobot