yryun

I'm using matlab 2015b and 2017a. In 2017a, it is okay to recv much data. but in 2015b, over 10,000 channel data recv make it shutdown. is there any solution...

Too much VRAM cost and ckpt file volume compared to the paper

1

Thanks for great work! I followed the example provided with cfill dataset. -> https://github.com/sled-group/CycleNet/blob/main/docs/train.md I found 2 differences with the original works. 1. Training with batch size = 1 cost...