It may be just me, but I'm trying to learn MXNet.jl by doing, given that I can't find many examples around. I have the following code
a = [1 2 3 4;
4 3 2 1]
b = [4 3 2 1;
1 2 3 4]
mx.Correlation(mx.NDArray(reshape(a, (1,1,2,4))), mx.NDArray(reshape(b, (1,1,2,4))))
that gives me this error:
Check failed: top_width_ >= 1U (0 vs. 1) Correlation cannot be done with current settings.Neighborhood and kernel don't fit in blob
I tried looking for that in the source code, but I struggle to extrapolate what that condition means. Can anyone shed some light on this?
It may be just me, but I'm trying to learn
MXNet.jlby doing, given that I can't find many examples around. I have the following codethat gives me this error:
Check failed: top_width_ >= 1U (0 vs. 1) Correlation cannot be done with current settings.Neighborhood and kernel don't fit in blobI tried looking for that in the source code, but I struggle to extrapolate what that condition means. Can anyone shed some light on this?