5 ms·

This is an awesome resource that I keep coming back to again and again. Save this somewhere on your computer so you'll have it handy whenever you see some weird

• highd 9y ago
  I used to use this a bunch when calculating gradient descent algorithms, but I've since moved to using libraries with auto-differentiation (numpy with autodiff or theano) and it's a huge force multiplier for my work (not to mention GPU implementations basically for free). Of course there's always some derivation that's easiest done by hand.
  • netheril96 9y ago
    Isn't tensorflow the most popular auto differential library these days?
    • gtani 9y ago
      Probably you could look at # Watches, Stars, forks etc to gauge that, but if you want to look at vs theano, torch, pyTorch, Caffe2, start with cs231N lecture: http://cs231n.stanford.edu/slides/2017/cs231n_2017_lecture8.pdf http://cs231n.stanford.edu/slides/2017/cs231n_2017_lecture8.... (I think there's vid on youtube)