BPR#

class libreco.algorithms.BPR(task='ranking', data_info=None, loss_type='bpr', embed_size=16, norm_embed=False, n_epochs=20, lr=0.001, lr_decay=False, epsilon=1e-05, reg=None, batch_size=256, sampler='random', num_neg=1, use_tf=True, seed=42, lower_upper_bound=None, tf_sess_config=None, optimizer='adam', num_threads=1)[source]#

Bases: EmbedBase

Bayesian Personalized Ranking algorithm.

BPR is implemented in both TensorFlow and Cython.

Caution

  • BPR can only be used in ranking task.

  • BPR can only use bpr loss in loss_type.

Parameters:

  • task ({'ranking'}) – Recommendation task. See Task.

  • data_info (DataInfo object) – Object that contains useful information for training and inference.

  • loss_type ({'bpr'}) – Loss for model training.

  • embed_size (int, default: 16) – Vector size of embeddings.

  • norm_embed (bool, default: False) – Whether to l2 normalize output embeddings.

  • n_epochs (int, default: 10) – Number of epochs for training.

  • lr (float, default 0.001) – Learning rate for training.

  • lr_decay (bool, default: False) – Whether to use learning rate decay.

  • epsilon (float, default: 1e-5) – A small constant added to the denominator to improve numerical stability in Adam optimizer. According to the official comment, default value of 1e-8 for epsilon is generally not good, so here we choose 1e-5. Users can try tuning this hyperparameter if the training is unstable.

  • reg (float or None, default: None) – Regularization parameter, must be non-negative or None.

  • batch_size (int, default: 256) – Batch size for training.

  • sampler ({'random', 'unconsumed', 'popular'}, default: 'random') –

    Negative sampling strategy.

    • 'random' means random sampling.

    • 'unconsumed' samples items that the target user did not consume before.

    • 'popular' has a higher probability to sample popular items as negative samples.

    New in version 1.1.0.

  • num_neg (int, default: 1) – Number of negative samples for each positive sample, only used in ranking task.

  • use_tf (bool, default: True) – Whether to use TensorFlow or Cython version. The TensorFlow version is more accurate, whereas the Cython version is faster.

  • seed (int, default: 42) – Random seed.

  • lower_upper_bound (tuple or None, default: None) – Lower and upper score bound for rating task.

  • tf_sess_config (dict or None, default: None) – Optional TensorFlow session config, see ConfigProto options.

  • optimizer ({'sgd', 'momentum', 'adam'}, default: 'adam') – Optimizer used in Cython version.

  • num_threads (int, default: 1) – Number of threads used in Cython version.

References

Steffen Rendle et al. BPR: Bayesian Personalized Ranking from Implicit Feedback.

fit(train_data, neg_sampling, verbose=1, shuffle=True, eval_data=None, metrics=None, k=10, eval_batch_size=8192, eval_user_num=None, num_workers=0)[source]#

Fit BPR model on the training data.

Parameters:

  • train_data (TransformedSet object) – Data object used for training.

  • neg_sampling (bool) –

    Whether to perform negative sampling for training or evaluating data.

    New in version 1.1.0.

  • verbose (int, default: 1) – Print verbosity. If eval_data is provided, setting it to higher than 1 will print evaluation metrics during training.

  • shuffle (bool, default: True) – Whether to shuffle the training data.

  • eval_data (TransformedSet object, default: None) – Data object used for evaluating.

  • metrics (list or None, default: None) – List of metrics for evaluating.

  • k (int, default: 10) – Parameter of metrics, e.g. recall at k, ndcg at k

  • eval_batch_size (int, default: 8192) – Batch size for evaluating.

  • eval_user_num (int or None, default: None) – Number of users for evaluating. Setting it to a positive number will sample users randomly from eval data.

  • num_workers (int, default: 0) – How many subprocesses to use for data loading. 0 means that the data will be loaded in the main process.

get_item_embedding(item=None, include_bias=False)#

Get item embedding(s) from the model.

Parameters:

  • item (int or str or None, default: None) – Query item id. If it is None, all item embeddings will be returned.

  • include_bias (bool, default: False) – Whether to include bias term in returned embeddings.

Returns:

item_embedding – Returned item embeddings.

Return type:

numpy.ndarray

Raises:

  • ValueError – If the item does not appear in the training data.

  • AssertionError – If the model has not been trained.

get_user_embedding(user=None, include_bias=False)#

Get user embedding(s) from the model.

Parameters:

  • user (int or str or None, default: None) – Query user id. If it is None, all user embeddings will be returned.

  • include_bias (bool, default: False) – Whether to include bias term in returned embeddings.

Returns:

user_embedding – Returned user embeddings.

Return type:

numpy.ndarray

Raises:

  • ValueError – If the user does not appear in the training data.

  • AssertionError – If the model has not been trained.

init_knn(approximate, sim_type, M=100, ef_construction=200, ef_search=200)#

Initialize k-nearest-search model.

Parameters:

  • approximate (bool) – Whether to use approximate nearest neighbor search. If it is True, nmslib must be installed. The HNSW method in nmslib is used.

  • sim_type ({'cosine', 'inner-product'}) – Similarity space type.

  • M (int, default: 100) – Parameter in HNSW, refer to nmslib doc.

  • ef_construction (int, default: 200) –

    Parameter in HNSW, refer to nmslib doc.

  • ef_search (int, default: 200) –

    Parameter in HNSW, refer to nmslib doc.

Raises:

  • ValueError – If sim_type is not one of (‘cosine’, ‘inner-product’).

  • ModuleNotFoundError – If approximate=True and nmslib is not installed.

classmethod load(path, model_name, data_info, **kwargs)#

Load saved embed model for inference.

Parameters:

  • path (str) – File folder path to save model.

  • model_name (str) – Name of the saved model file.

  • data_info (DataInfo object) – Object that contains some useful information.

Returns:

model – Loaded embed model.

Return type:

type(cls)

See also

save

predict(user, item, cold_start='average', inner_id=False)#

Make prediction(s) on given user(s) and item(s).

Parameters:

  • user (int or str or array_like) – User id or batch of user ids.

  • item (int or str or array_like) – Item id or batch of item ids.

  • cold_start ({'popular', 'average'}, default: 'average') –

    Cold start strategy.

    • ’popular’ will sample from popular items.

    • ’average’ will use the average of all the user/item embeddings as the representation of the cold-start user/item.

  • inner_id (bool, default: False) – Whether to use inner_id defined in libreco. For library users inner_id may never be used.

Returns:

prediction – Predicted scores for each user-item pair.

Return type:

float or numpy.ndarray

rebuild_model(path, model_name, full_assign=True)#

Assign the saved model variables to the newly initialized model.

This method is used before retraining the new model, in order to avoid training from scratch every time we get some new data.

Parameters:

  • path (str) – File folder path for the saved model variables.

  • model_name (str) – Name of the saved model file.

  • full_assign (bool, default: True) – Whether to also restore the variables of Adam optimizer.

recommend_user(user, n_rec, cold_start='average', inner_id=False, filter_consumed=True, random_rec=False)#

Recommend a list of items for given user(s).

Parameters:

  • user (int or str or array_like) – User id or batch of user ids to recommend.

  • n_rec (int) – Number of recommendations to return.

  • cold_start ({'popular', 'average'}, default: 'average') –

    Cold start strategy.

    • ’popular’ will sample from popular items.

    • ’average’ will use the average of all the user/item embeddings as the representation of the cold-start user/item.

  • inner_id (bool, default: False) – Whether to use inner_id defined in libreco. For library users inner_id may never be used.

  • filter_consumed (bool, default: True) – Whether to filter out items that a user has previously consumed.

  • random_rec (bool, default: False) – Whether to choose items for recommendation based on their prediction scores.

Returns:

recommendation – Recommendation result with user ids as keys and array_like recommended items as values.

Return type:

dict of {Union[int, str, array_like] : numpy.ndarray}

save(path, model_name, inference_only=False, **kwargs)#

Save embed model for inference or retraining.

Parameters:

  • path (str) – File folder path to save model.

  • model_name (str) – Name of the saved model file.

  • inference_only (bool, default: False) – Whether to save model only for inference. If it is True, only embeddings will be saved. Otherwise, model variables will be saved.

See also

load

search_knn_items(item, k)#

Search most similar k items.

Parameters:

  • item (int or str) – Query item id.

  • k (int) – Number of similar items.

Returns:

similar items – A list of k similar items.

Return type:

list

search_knn_users(user, k)#

Search most similar k users.

Parameters:

  • user (int or str) – Query user id.

  • k (int) – Number of similar users.

Returns:

similar users – A list of k similar users.

Return type:

list