My Python Code for the Netflix Prize

Here's mine if anyone is interested. I wrote it in D and haven't looked at it in years. I'm sure it's not usable as-is, but it might be fun anyway.<p><a href="https://github.com/nogwater/NetflixPrizeD" rel="nofollow">https://github.com/nogwater/NetflixPrizeD</a><p>The algorithm is based on Simon Funk's blog post here: <a href="http://sifter.org/~simon/journal/20061211.html" rel="nofollow">http://sifter.org/~simon/journal/20061211.html</a><p>For me, the best part was squeezing the data and indexes into memory. :)

@tuananh I've got the dataset stored away, but I don't know if I'm legally allowed to post it. Would love if someone could produce proof one way or the other.<p>@viraj_shah I spent about 6 months working on the project before I had to stop to concentrate on my schoolwork (I was a senior in collge at the time). I think it would have been impossible to do this for myself without Cython. If it were to happen today, I would probably be writing in PyCuda, or with Numba, and it would be much, much, MUCH more succinct.

I wrote a 195-page monograph on the Netflix Prize, for people interested in that sort of stuff: <a href="http://arek-paterek.com/book" rel="nofollow">http://arek-paterek.com/book</a>

You indent by 8 characters.... I wanted to read your code but this will make my eyes bleed.<p>From pep 8: "Use 4 spaces per indentation level."

Very nice. It might be helpful to (briefly) describe the actual techniques you tried in the readme file? At least that's the first thing I looked for...

This was incredibly kind of you to post up. It is great to see it public domain as many can learn from it. The Cython code looks scary though - 18k lines! May I ask how long you spent on this?

Nitpick: binary blobs like .pyc and .so don't belong in a code repository. Instead you would put a makefile or setup.py to compile the .pyx files.

I also worked on this at uni and had lots of fun -- those lessons certainly look familiar! We were trying to mine Wikipedia for more information on the movies. The code's here:<p><a href="http://code.google.com/p/wikipedia-netflix/wiki/WikipediaNetflix" rel="nofollow">http://code.google.com/p/wikipedia-netflix/wiki/WikipediaNet...</a><p>It includes Wikipedia parsing stuff and a fairly fast C++ implementation of the very cool BellKor kNN algorithm.

Thanks for posting this. Your code combines two successful approachs : a latent factor model (SVD) and a neighborhood model :)<p>Here's my implementation of a recommender algorithm in C if someone is interested : <a href="https://github.com/GHamrouni/Recommender" rel="nofollow">https://github.com/GHamrouni/Recommender</a>

Mine is at <a href="https://github.com/hbcdev/Netflix" rel="nofollow">https://github.com/hbcdev/Netflix</a> in C++, got to the top 500. It runs just inside my 8GB PC :)

I had a look though this and was confused by this line:<p>ratings[ratings&#60;1.0] = 1.0<p>in<p><a href="https://github.com/alexbw/Netflix-Prize/blob/master/src/predict.py" rel="nofollow">https://github.com/alexbw/Netflix-Prize/blob/master/src/pred...</a><p>Is it specific to NumPy? Or perhaps a Python trick I haven't seen before?

great for ref. however i can't found the dataset anywhere on the internet :(

Good readme file. Liked the lessons learned.

Does anyone have any thoughts on the lack of conformity to PEP8? His code works and it's valid python but I feel that it's difficult to read.