| Safe Haskell | None |
|---|---|
| Language | Haskell2010 |
Numeric.Datasets.Netflix
Description
Netflix prize dataset
From the README : The movie rating files contain over 100 million ratings from 480 thousand randomly-chosen, anonymous Netflix customers over 17 thousand movie titles. The data were collected between October, 1998 and December, 2005 and reflect the distribution of all ratings received during this period. The ratings are on a scale from 1 to 5 (integral) stars. To protect customer privacy, each customer id has been replaced with a randomly-assigned id. The date of each rating and the title and year of release for each movie id are also provided.
The competition ended on September, 2009, and the dataset was subsequently removed from the public domain by the company (see http://netflixprize.com/).
We include in the repository a tiny subset of the original dataset for development purposes. Since we use `file-embed` to load the data, the directories are hardcoded (see the Datasets section below); users may either symlink or copy the full dataset in the given directories.
- parseTrainingSet :: Num a => Either String [(UserId, MovieId, RD a)]
- parseTestSet :: Either String [(UserId, MovieId, Day)]
- parseMovies :: Either String [Movie]
- data RD a = RD {}
- data UserId
- data MovieId
- data Train = Train {
- trainRating :: RatingDate
- rating :: Int
- newtype Test = Test {}
- data Movie = Movie {
- movieId :: MovieId
- releaseYear :: Day
- movieTitle :: ByteString
- data RatingDate = RatingDate {
- userId :: UserId
- ratingDate :: Day
- trainingSet :: [(FilePath, ByteString)]
- testSet :: [(FilePath, ByteString)]
- movies :: [(FilePath, ByteString)]
Dataset parsing and shaping
parseTrainingSet :: Num a => Either String [(UserId, MovieId, RD a)] Source #
Parse the whole training set, convert to coordinate format and concatenate into a single list.
parseTestSet :: Either String [(UserId, MovieId, Day)] Source #
Parse the whole test set, convert to coordinate format and concatenate into a single list.
parseMovies :: Either String [Movie] Source #
Parse the whole movies file, convert to coordinate format and concatenate into a single list.
Types
A type for date-tagged movie ratings
Training set item
Constructors
| Train | |
Fields
| |
Test set item
Constructors
| Test | |
Fields | |
Movie dataset item
Constructors
| Movie | |
Fields
| |
data RatingDate Source #
A date-tagged movie rating
Constructors
| RatingDate | |
Fields
| |
Instances
Datasets
trainingSet :: [(FilePath, ByteString)] Source #
The training set (a set of text files) is assumed to be in the directory `datafiles/netflix/training/` relative to the repository root
testSet :: [(FilePath, ByteString)] Source #
The test set (one text file) is assumed to be in the directory `datafiles/netflix/test/` relative to the repository root
movies :: [(FilePath, ByteString)] Source #
The movies dataset (one text file) is assumed to be in the directory `datafiles/netflix/movies/` relative to the repository root