boltzmannentropy / interviews.ai Goto Github PK

The answer in SOL-140 Number 2 should be "Sampling with replacement". If you sample with replacement the samples should be independent and have zero covariance. This is backed up by the content at this page.

Wanted to flag this in case it hasn't been seen yet. I think the solutions for SOL-208  CH.SOL- 8.32 should be -0.165 and for 8.51, -0.164. I may be wrong but wanted to run it by you.

From a reader:
Because solutions are not immediately below the questions there was a lot of tab switching on my laptop. I would also prefer multiple questions inside a single problem to be rewritten as separate problems. In general, I prefer reading textbooks like this in the following order:
question1 -> answer1 -> question2 -> answer2 -> question3 -> pausing and thinking on answer -> answer3 -> ...

errata: pg43 PRB-31 s/bare/bear - the object will bear a specific characteristic, bare is not the right word.

First of all, Thank You Very Much for open sourcing this amazing book.

As per the issue, I downloaded the the book through which I managed to get a copy of the book which was a pdf file named "deep-learning-interviews.pdf".

But when I tried to read the pdf on the iBook my iPad mini, it came up as PDF password protected and I was not able to access the pdf.

Is there a way to alleviate this issue?

Thank in advance.

When is the volume 2 of the book expected to be released?

Hi,
Very interesting book indeed, so thanks for the hard work!

I noticed a quick typo while skimming through the book:

• page 267: Hi realized two consecutive layers in PyTorch [20], declared as follows (8.3):
  • Hi should probably be He

Thanks!

There is an error in Python code solution CH.SOL- 5.21.
The sigmoid derivative is y * (1 - y) not x * (1 - x).

Love the book! Ran into this display issue when viewing it on Chrome's PDF viewer: