RuntimeWarning: Iterating over a tensor might cause the trace to be incorrect

[vincentily]

when I run atuotvm for my net, I met this warning , does anybody know how to solve it

[masahi]

This is coming from PyTorch, not TVM.