Skip to first unread message
julius...@gmail.com
Oct 30, 2016, 10:13:05 PM10/30/16
to Caffe Users
I am trying to train a LRCN model for image caption. https://github.com/jeffdonahue/caffe/tree/recurrent/examples/coco_caption
I have run the train_lrcn.vgg.sh , but the final loss is ~47, which seems quit high.
I find several training scripts in the file and I have no idea that what is the right step to train a LRCN model for coco caption ?
Reply all
Reply to author
Forward
0 new messages