[FCN] problem with train.prototxt in voc-fcn32s

[Xubin LI]

https://github.com/shelhamer/fcn.berkeleyvision.org/tree/master/voc-fcn32s

train.prototxt generated by net.py is not same with original train.prototxt.

The  difference is the dropout layer.  is there any difference in result?

[Evan Shelhamer]

You should follow the definition made by net.py. I'll correct the included definition in the repo to reflect this.

For what it's worth, in our experiments the results were nearly the same with or without dropout.

Evan Shelhamer

--
You received this message because you are subscribed to the Google Groups "Caffe Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to caffe-users+unsubscribe@googlegroups.com.
To post to this group, send email to caffe...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/caffe-users/d784fbe2-caf2-4c01-b737-dd558922800a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.