kelayamatoz

Thanks a lot for your advice; the graph is quite self-explanatory and the visualization looks really nice. I will try a few scale-free networks and will let you know if...

It seems that our LUT template is implemented as a list of registers instead of a chisel ROM, and that would very likely prevent the downstream synthesizer to recognize the...

Hi David, My name is Tian. I'm moving our discussion from email to this issue ticket so that other developers who have issues with this error could see it. The...

It seems that the ExponentailMovingAverage doesn't help with the training processs. Disabling the related lines would solve this issue. After resolving this issue, I also discovered a similar reuse-variable issue...

Hey @distantJing: Check this one: https://github.com/kelayamatoz/bi-att-flow-lstm-extractor/blob/master/basic/model.py#L25-L36

Essentially you need to put the exponential smoothing variables into a different scope and make sure that each GPU gets a unique set of loss variables.