tfrs.examples.movielens.sample_listwise

Function for converting the MovieLens 100K dataset to a listwise dataset.

tfrs.examples.movielens.sample_listwise(
    rating_dataset: tf.data.Dataset,
    num_list_per_user: int = 10,
    num_examples_per_list: int = 10,
    seed: Optional[int] = None
) -> tf.data.Dataset

Used in the notebooks

Used in the tutorials

rating_dataset The MovieLens ratings dataset loaded from TFDS with features "movie_title", "user_id", and "user_rating".
num_list_per_user An integer representing the number of lists that should be sampled for each user in the training dataset.
num_examples_per_list An integer representing the number of movies to be sampled for each list from the list of movies rated by the user.
seed An integer for creating np.random.RandomState.

A tf.data.Dataset containing list examples.

Each example contains three keys: "user_id", "movie_title", and "user_rating". "user_id" maps to a string tensor that represents the user id for the example. "movie_title" maps to a tensor of shape [sum(num_example_per_list)] with dtype tf.string. It represents the list of candidate movie ids. "user_rating" maps to a tensor of shape [sum(num_example_per_list)] with dtype tf.float32. It represents the rating of each movie in the candidate list.