FlexMF BPR

This page analyzes the hyperparameter tuning results for the FlexMF scorer in implicit-feedback mode with pairwise loss (Bayesian Personalized Ranking).

Parameter Search Space

/home/mde48/lenskit/lenskit-codex/.venv/lib/python3.12/site-packages/ray/tune/search/sample.py:700: RayDeprecationWarning: The `base` argument is deprecated. Please remove it as it is not actually needed in this method.

Parameter	Type	Distribution	Values	Selected
embedding_size	Integer	LogUniform	4 ≤ \(x\) ≤ 512	38
regularization	Float	LogUniform	0.0001 ≤ \(x\) ≤ 10	0.033
learning_rate	Float	LogUniform	0.001 ≤ \(x\) ≤ 0.1	0.00414
reg_method	Categorical	Uniform	L2, AdamW	AdamW
negative_count	Integer	Uniform	1 ≤ \(x\) ≤ 5	4
item_bias	Categorical	Uniform	True, False	False

Final Result

Searching selected the following configuration:

{
    'embedding_size': 38,
    'regularization': 0.03301715243697084,
    'learning_rate': 0.004143227701688067,
    'reg_method': 'AdamW',
    'negative_count': 4,
    'item_bias': False,
    'epochs': 11
}

With these metrics:

{
    'RBP': 0.21577414266433792,
    'LogRBP': 2.26071691560203,
    'NDCG': 0.4529772896895511,
    'RecipRank': 0.40095257811434615,
    'TrainTask': 'a8e23294-bef8-4b5c-9c88-9208ba12a8f5',
    'TrainTime': None,
    'TrainCPU': None,
    'max_epochs': 50,
    'done': False,
    'training_iteration': 11,
    'trial_id': 'c9cea786',
    'date': '2025-05-06_00-00-09',
    'timestamp': 1746504009,
    'time_this_iter_s': 20.3419086933136,
    'time_total_s': 246.0682191848755,
    'pid': 442888,
    'hostname': 'CCI-ws21',
    'node_ip': '10.248.127.152',
    'config': {
        'embedding_size': 38,
        'regularization': 0.03301715243697084,
        'learning_rate': 0.004143227701688067,
        'reg_method': 'AdamW',
        'negative_count': 4,
        'item_bias': False,
        'epochs': 11
    },
    'time_since_restore': 246.0682191848755,
    'iterations_since_restore': 11
}