Hacker News Reader

I saw what you wrote about your dot product speed issue. Did you try using NumPy's einsum function? http://docs.scipy.org/doc/numpy/reference/generated/numpy.ei...

It's really fast for this kind of stuff. Happy to give details about how to use it if you need.

View on HN

Hey SandB0x, thanks for the advice! I actually started off by using numpy.dot, which is precisely what's needed. The problem is that I need it to go even faster (currently takes a few seconds) but this function is already heavily optimized and uses Intel MKL to accelerate the math. In fact, my cython implementation would be slower than numpy.dot were it not for some embedded logic that breaks out of the dot product half way through calculation. As I'm computing the row * row, going element by element, if the running sum of those products gets to be really negative (indicating that in the dimensions multiplied so far, the two rows are highly dissimilar) I stop tryin to calculate the rest of the row, thereby saving me from calculating the full dot product. This is cheating obvioisly, but it's x3 or x4 faster numpy.dot. So, because there's branching logic in my implementation of the dot product, I can't express it interns of Einstein summations.

Reply on news.ycombinator.com