Chap15 conv1d and conv2d about python-machine-learning-book-2nd-edition HOT 3 OPEN

Had a look at the Jupyter Notebook of 3rd Edition
conv1d:
the code was changed to
3rd edition: for i in range(0, int((len(x_padded) - len(w_rot)) / s) + 1, s):
when you use this code with example in Out[5] it returns the wrong result: Conv1d Implementation p=0 s=2: [7. 9.]
i think the "/ s" is wrong, as you take s as step

conv2d:
the code was not changed.
example in Out[11] still results in [[4.6]]

I only compared the for-statements. If there were further changes in the code, i did not check this.

from python-machine-learning-book-2nd-edition.

You are right. Arg, yeah, the /s shouldn't be there. Probably didn't notice due to setting s=1. Need to check this more thoroughly and update it. I will add a reference on the 3rd editions issue tracker so that it can be fixed in the 4th edition.

from python-machine-learning-book-2nd-edition.

Thanks for the note. It's been a few years ago, but I vaguely remember this issue. I believe we had fixed that in the 3rd edition: https://github.com/rasbt/python-machine-learning-book-3rd-edition/blob/master/ch15/ch15_part1.ipynb

from python-machine-learning-book-2nd-edition.