blankWorld / GroupNorm-caffe

caffe implementation of Group Normalization https://arxiv.org/abs/1803.08494
41 stars 8 forks source link

Did you train the ResNet model with this GroupNorm? #2

Open youngwanLEE opened 5 years ago

youngwanLEE commented 5 years ago

Did you train or check the performance compared to original paper?

yuzcccc commented 5 years ago

I am also interested in this question, does any one reimpl. the results in original paper? Or could you please provide an complete prototxt to explain how to use this layer in Faster-R-CNN model @blankWorld . Thanks a lot!