FlexMF Explicit

This page analyzes the hyperparameter tuning results for the FlexMF scorer in explicit-feedback mode (a biased matrix factorization model trained with PyTorch).

Parameter Search Space

Parameter Type Distribution Values Selected
embedding_size Integer LogUniform 4 ≤ \(x\) ≤ 512 217
regularization Float LogUniform 0.0001 ≤ \(x\) ≤ 10 2.28
learning_rate Float LogUniform 0.001 ≤ \(x\) ≤ 0.1 0.0287
reg_method Categorical Uniform L2, AdamW AdamW

Final Result

Searching selected the following configuration:

{
    'embedding_size': 217,
    'regularization': 2.280494473138675,
    'learning_rate': 0.02874941091422996,
    'reg_method': 'AdamW',
    'epochs': 6
}

With these metrics:

{
    'RBP': 0.02801595182600994,
    'LogRBP': 0.21925874681457902,
    'NDCG': 0.21458613490289732,
    'RecipRank': 0.11385921973095289,
    'RMSE': 0.9291464572861081,
    'TrainTask': '17de94b2-e420-4444-b777-245150b7d05f',
    'TrainTime': None,
    'TrainCPU': None,
    'max_epochs': 50,
    'done': False,
    'training_iteration': 6,
    'trial_id': 'd5514e0d',
    'date': '2025-05-06_22-47-11',
    'timestamp': 1746586031,
    'time_this_iter_s': 0.5432236194610596,
    'time_total_s': 3.3543996810913086,
    'pid': 870366,
    'hostname': 'gracehopper1',
    'node_ip': '192.168.225.60',
    'config': {
        'embedding_size': 217,
        'regularization': 2.280494473138675,
        'learning_rate': 0.02874941091422996,
        'reg_method': 'AdamW',
        'epochs': 6
    },
    'time_since_restore': 3.3543996810913086,
    'iterations_since_restore': 6
}

Parameter Analysis

Embedding Size

The embedding size is the hyperparameter that most affects the model’s fundamental logic, so let’s look at performance as a fufnction of it:

Learning Parameters

Iteration Completion

How many iterations, on average, did we complete?

How did the metric progress in the best result?

How did the metric progress in the longest results?