Dmitrij Koniajev

Same for me

@iamtrask but in fact, the code proposed in the book seems difficult to understand. I think it would be much clearer to write it just like this: ``` weight_deltas =...

> I second this. The weight_delta matrix should have the same number of columns as the length of the "input" vector. It should be > > weight_deltas = outer_prod(delta,input). @PeterGanZW...

Thanks for the implementation in the Rast language. You also noticed incorrectly calculated weights in Chapter 5: https://github.com/suyash/grokking-deep-learning-rs/blob/master/examples/chapter5.rs#L215 I've registered issue for this: #23