The Author Online Book Forums are Moving

The Author Online Book Forums will soon redirect to Manning's liveBook and liveVideo. All book forum content will migrate to liveBook's discussion forum and all video forum content will migrate to liveVideo. Log in to liveBook or liveVideo with your Manning credentials to join the discussion!

Thank you for your engagement in the AoF over the years! We look forward to offering you a more enhanced forum experience.

ssharma (51) [Avatar] Offline
#1
Dear Author,

I am enjoying your book but cant follow along with lack of examples data and code. Could I request you to post the same? Thanks in advance.

Cheers,
ssharma (51) [Avatar] Offline
#2
appreciate any response here. thanks,
ssharma (51) [Avatar] Offline
#3
followup.
ssharma (51) [Avatar] Offline
#4
Looks like it is a dead forum?
henrik.brink (22) [Avatar] Offline
#5
Hi there,

I'm terribly sorry for the lack of response! We have lost all notifications from the forums after the software was updated it seems, so we had no idea new messages were waiting. It's no excuse for not checking for so long, so again, very sorry.

To try to answer your question:

Yes, in addition to the reamining chapters we are actually working on collecting all the coding and data samples and make them more accessible. This will be part of one of the coming MEAP updates.

Again, I'm terribly sorry for the delay in responding. Thank you for your feedback and hope you will continue to be involved.

- Henrik

ssharma (51) [Avatar] Offline
#6
Thanks Henrik. Appreciate any update on code/sample data.
332471 (1) [Avatar] Offline
#7
The code would be super helpful.

With the way the book is laid out, maybe you could just give Ipython notebooks for each project: Titanic, German Credit Data, etc.

I love the content and real world examples that are explained here. Personally I've been going through the Titanic data set and I can work through the documentation on scikit-learn and stackoverflow to understand how you get a nonlinear decision boundary plot shown on Figure 3.8... but it's a fair amount of code to get from point A to B

Great book!
ssharma (51) [Avatar] Offline
#8
Should I hope to have some code in the next MEAP update?
aluna (2) [Avatar] Offline
#9
Authors said they will provide the code. And it is available as of today.
ssharma (51) [Avatar] Offline
#10
Thanks for the update but I do not see any link to download the code? I looked at the book page as well as "account.manning.com".
aluna (2) [Avatar] Offline
#11
ah so sorry.
The code was from another book I have in my account :-/
Sorry for the wrong info.
ssharma (51) [Avatar] Offline
#12
any luck with the code & data?
ssharma (51) [Avatar] Offline
#13
any luck?
6960 (3) [Avatar] Offline
#14
So any updates now, regarding the code samples?

Iphyton notebooks would be much appreciated.
ssharma (51) [Avatar] Offline
#15
.
ssharma (51) [Avatar] Offline
#16
+1
henrik.brink (22) [Avatar] Offline
#17
Hi!

We're collecting code and data samples in IPython notebooks here:

https://github.com/brinkar/real-world-machine-learning

There are currently examples available for 2 chapters, but the rest should show up in the coming week. Please let us know if you have any questions regarding this format and we'll do our best to accomodate.

Thanks again for all your patience and support smilie

- Henrik
86409 (3) [Avatar] Offline
#18
Hi Henrik,

Really enjoying your book so far and very happy the code supplement is now available. I am not of a strong math background and I have put down many a machine learning book because they lost me in the maths; this book however is starting out to be just what I needed, a practical introduction on how to use machine learning algorithms. Hopefully from here I can build my math understanding while actually being able to do some useful machine learning activities.

I'm a bit new to Python but I saw in the Chapter 3 - Modeling and prediction.ipynb file that when the data is split 80/20

# We make a 80/20% train/test split of the data
data_train = data[:int(0.8*len(data))]
data_test = data[int(0.8*len(data)):]


which looks like its taking the last 80% of the data for the data_train and the first 80% of the data for data_test. Should this be
more like

# We make a 80/20% train/test split of the data
data_train = data[:int(0.8*len(data))]
data_test = data[int(0.2*len(data)):]


Or am I misunderstanding how this code is working?

Thanks for your dedication to creating this great book.

Greg
henrik.brink (22) [Avatar] Offline
#19
Hi Greg,

Thanks so much for your support and for your question.

This code can be a bit confusing because of how Python lists and ranges work, but I believe it's correct. The ":" sign means "from the beginning" or "to the end" in a range, depending on which side of the number it is. Here's a simple IPython session to test it:

In [7]: a = [0,1,2,3,4,5,6,7,8,9]

In [8]: a[:7]
Out[8]: [0, 1, 2, 3, 4, 5, 6]

In [9]: a[7:]
Out[9]: [7, 8, 9]


Does that make sense?

BTW, for others following this post as well, there should now be code available for all chapters (at least those that have code in them) in the Github repo. Looking forward to hearing your feedback!

- Henrik
86409 (3) [Avatar] Offline
#20
Hi Henrik,

The example was very helpful, I see now.

The first line is taking 0-79% and the 2nd line it taking 80-100% giving us the 80/20 split.

Thank you very much!

Greg
255806 (2) [Avatar] Offline
#21
The book description tells that it will have Python and R code, but there are only Python code on github.

Will R code inserted in the near future or just Python will be available?


Tks,

Ícaro
ssharma (51) [Avatar] Offline
#22
+1
290422 (2) [Avatar] Offline
#23
What about your own words: "Code examples are in Python and R. No prior machine learning experience required."?

I hope you are going to keep your promise.
429451 (1) [Avatar] Offline
#24
Once you know R, using Python and Jupyter notebooks or Spyder in Anaconda is pretty easy to learn. I personally think Python is probably better/easier to use for analysis of unstructured data than R (but R is catching up thanks to the folks at RStudio like Hadley Wickham). I think it is well worth the effort to install Anaconda and run the ipython notebooks (I'm a big believer in using the best tool for the job smilie.

Good luck, Brian
drpositron (9) [Avatar] Offline
#25
R code for the book is available at https://padamson.github.io/rwml-R/