The errata list is a list of errors and their corrections that were found after the product was released.
The following errata were submitted by our customers and have not yet been approved or disproved by the author or editor. They solely represent the opinion of the customer.
Color Key: Serious technical mistake Minor technical mistake Language or formatting error Typo Question Note Update
| Version |
Location |
Description |
Submitted by |
Date submitted |
| Printed, Other Digital Version |
Page ii
Revision History for the First Edition |
I just bought this book from an Amazon new/used dealer. The latest release mentioned is 2014-08-22 Third release. The latest release shown on the copyright page at Amazon is 2016-06-13.
My copy mentions that I can get release details at
http://www.oreilly.com/catalog/errata.cap?isbn=9781449370787
I get an error trying to go there.
At http://www.oreilly.com/catalog/errata.csp?isbn=0636920030720
I cannot find any details about the 4th release.
The preface in my book is different in some important details from the sample pages at Amazon. E.g. the mention of the continuum version of Python is missing.
What else is different? I am wondering if I should return the book I have, and buy the latest release.
|
James Jennings |
Aug 30, 2016 |
| PDF |
Page 7
M&M Table |
The table is written as:
Prior p(H) | Likelihood p(D|H) | p(H) p(D|H) | Posterior p(H|D)
A 1/2 | (20)(20) | 200 | 20/27
B 1/2 | (10)(14) | 70 | 7/27
but it should be:
Prior p(H) | Likelihood p(D|H) | p(H) p(D|H) | Posterior p(H|D)
A 1/2 | 20/100 * 20/100 | 1/2 * 20/100 * 20/100 | 20/27
B 1/2 | 10/100 * 14/100 | 1/2 * 10/100 * 14/100 | 7/27
|
João Marcos Gris |
Sep 08, 2017 |
| Printed |
Page 7
Table |
In the third column of the table, the likelihood p(D|H) is computed as (20)(20) for bag A and (10)(14) for bag B.
Should it not be (20)(10) for bag A and (20)(14) for bag B?
|
Anonymous |
Dec 06, 2017 |
| Printed |
Page 20
Second set of results |
The values reported are wrong for the second set of exercise with data.
The correct results are:
4 0.0
6 0.0
8 0.9158452719690099
12 0.08040342579700499
20 0.0037513022339850668
and not:
4 0.0
6 0.0
8 0.932....
12 0.0522..
20 0.0015...
|
Anonymous |
May 05, 2021 |
| Printed |
Page 48
United States |
Below the line that starts with "First we create a Pmf..." the mapping of each die to its probability doesn't seem to reflect the number of each type of die listed on page 47.
E.g shouldn't it be a 5 instead of a 2 in the line:
"pmf_dice.Set(Die(4),2)"
Either there is a mistake or the relationship between that list and the corresponding lines of code isn't made clear enough.
|
Erdem Karakoylu |
Dec 23, 2014 |
| Printed |
Page 48
4th paragraph |
Shouldn't it be CDF(x) = p(x<=X) instead of p(X<=x)?
|
Fabio |
May 12, 2019 |
| Printed |
Page 158
Second code snippet |
I believe the comment that says #class Detector should say #class Emitter as this method should be implemented for the Emitter.
|
Anonymous |
Feb 06, 2021 |
| Printed |
Page 160
Figure 14-2 |
Figure 14-2 shows the results of the hierchical model. However, the posteriors of r and n seem to have been mixed up in the legend.
As described in the section "Extracting the posteriors" on page 159, the posterior of the average rate r should be wider than the posterior of the actual rate n. This is what is expected intuitively since any uncertainty in r translates to at least as much uncertainty in n, which is mentioned in the text on page 159. Also, this is what I found when I replicated this exercise with homebrew numpy code (which I am happy to share).
In summary, the narrower curve with the highest peak should be r and the wider curve with the slightly lower peak is n.
The shapes of the curves are perfectly in order. Therefore the only necessary fix is to swap "posterior of r" and "posterior of n" in the legend.
|
Christiaan Erdbrink |
Apr 04, 2020 |