Recommendation

This module provides data types for recommendation and rating datasets.

These datasets are commonly used for collaborative filtering and recommendation system research.

Base Types

XPM Configdatamaestro_text.data.recommendation.RatedItems(*, id, ratings)

Bases: Base

id: str: The unique (sub-)dataset ID

ratings: datamaestro.data.File

Base class for datasets containing user ratings. The ratings attribute provides access to the ratings file.

MovieLens

XPM Configdatamaestro_text.data.recommendation.Movielens(*, id, ratings, links, movies, tags)

Bases: RatedItems

id: str: The unique (sub-)dataset ID

ratings: datamaestro.data.File

links: datamaestro.data.csv.Generic

movies: datamaestro.data.csv.Generic

tags: datamaestro.data.csv.Generic

MovieLens datasets include additional metadata:

ratings - User ratings (user_id, movie_id, rating, timestamp)
movies - Movie metadata (movie_id, title, genres)
tags - User-applied tags (user_id, movie_id, tag, timestamp)
links - Links to external sources (movie_id, imdb_id, tmdb_id)

Example usage:

from datamaestro import prepare_dataset

# Load MovieLens dataset
ml = prepare_dataset("io.grouplens.movielens.ml1m")

# Access ratings file
ratings_path = ml.ratings.path

# Access movie metadata
movies_path = ml.movies.path