None convergence with SoftmaxLoss when num_output becoming like 800k,

NVIDIA / caffe

Caffe: a fast open framework for deep learning.

http://caffe.berkeleyvision.org/

Other

672 stars 263 forks source link

None convergence with SoftmaxLoss when num_output becoming like 800k, #568

Open peyer opened 5 years ago

peyer commented 5 years ago

I have comment a problem while training a model for face recognition that when increasing num_output to 800k, the prob of each category became 1.0/800k and never channged! I have tried the same network and optimization parameters with BVLC-CAFFE, it successfully became to convergence after 8k iterations. How should I do ?