Errata

Generative Deep Learning

Errata for Generative Deep Learning, Second Edition

Submit your own errata for this product.

The errata list is a list of errors and their corrections that were found after the product was released. If the error was corrected in a later version or reprint the date of the correction will be displayed in the column titled "Date Corrected".

The following errata were submitted by our customers and approved as valid errors by the author or editor.

Color key: Serious technical mistake Minor technical mistake Language or formatting error Typo Question Note Update

Version Location Description Submitted By Date submitted Date corrected
Page Digital Version
Figure 9-7

Caption to Figure 9-7 mistakenly suggests that normalization statistics are calculated across blue cells, whereas normalization statistics are calculated across red cells.

Note from the Author or Editor:
Change the word blue to red in the caption

Mike Singer  Feb 22, 2024  Mar 01, 2024
Page Digital Version
Example 9-6

In Example 9-6, the line of code marked number 5, reads:
gpt = models.Model(inputs=inputs, outputs=[outputs, attention])

Here, "attention" should be "attention_scores" to be consistent with return variable names in line of code marked number 3.

Note from the Author or Editor:
change `attention` to `attention_scores` in Example 9-6 (on number 5 line)

Mike Singer  Feb 22, 2024  Mar 01, 2024
Figure 5-2 (of version of chap. 5 that is to be the 11th chapter of the final book)
Figure 5-2 (of version of chap. 5 that is to be the 11th chapter of the final book)

Just checking that Fig. 5-2 (of future chapter 11) does not have the "KEY" and "QUERY" labels flipped.

The corresponding paragraphs speak of "too" in the context of the query (making the query seem like a singular vector). And, they speak of the key as being "vectors" (plural). As such, the paragraphs had me expecting the figure to have "KEY" and "QUERY" to be flipped versus how they are shown.

I suspect that Foster has things correct and that I am wrong. Still, I hope to confirm that the figure is correct before I spend too much time convincing myself why the figure is correct.

Anonymous  Nov 05, 2022  Mar 07, 2023