Junru Jin

I found this error has something connection with parameters and separate constant value. If degree(including intermediate_graph_degree and graph_degree) sets to higher, it will be more correct. If `seperate_constant = torch.arange(batch_size).view(-1,...

I give a more easier testing code here: If print output are all zero, the results will be all in their boundary. ``` import torch import numpy as np import...

I found a more general case that CAGRA computation results will go wrong. Here, I use two different ways to calcualte the neighbors: 1) first way is to calculate all...

Same problem here with nni 3.0. But when use nni 2.5, the problem will disappear.

I check the nnictl_error.log and find that casued by this: `node:/lib64/libm.so.6: version 'GLIBC_2.27' not found (required by node)`. It seems I need to install GLIBC_2.27. But how can I address...

I write some code to visualize the activation function on each edge, but it seems not right. If somebody knows how to modify it, welcome to chat. 😄 ``` import...

> > > > I write some code to visualize the activation function on each edge, but it seems not right. If somebody knows how to modify it, welcome to...