getting loss of zero while training ssd 300x300 in kitti dataset

BVLC / caffe

Caffe: a fast open framework for deep learning.

http://caffe.berkeleyvision.org/

Other

34.09k stars 18.7k forks source link

Open dishita26 opened 5 years ago

dishita26 commented 5 years ago

Issue Summary caffe_issue image

I am not able to train ssd 300x300 in kitti dataset. The detection eval is zero while training.