cooganb

Thanks for this. It looks like PyTorch devteam removed implicit flattening of matrices for dot product, which is what causes this glitch. Here's the discussion: https://github.com/pytorch/pytorch/issues/2313

@dhpollack can you explain mm() ?

I'm wondering about speed, is there an easy way to invert one of the matrices? If it's two nx1 matrix, shouldn't that be a quick fix? On Sun, Oct 22,...

thanks so much for this snippet, super helpful to see. so, just to be clear, your method reshapes the matrix, then conducts batch matrix multiplication, correct? that's what I previously...

@dhpollack Last question, promise: is there a major difference between your solution and using `torch.squeeze(vector1).dot(torch.squeeze(vector2)` ?

@ricburton What's the status of this? Have you found someone to do the entire project?

@maxwellfoley3 Thank you for this! I'm having some trouble rebuilding it after specifying the new version. After running `npm install` then `npm run build`, I get: `Error: Cannot find module...

okay, I think that was just an install issue and think I've gotten past it. however, now I'm having issues with the adjusted file input format for 0.5.x compiler related...

You are indeed my code angel! Gracias. Any other code needed to run this?

@krofax @uF4No