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	182
regularization	Float	LogUniform	0.0001 ≤ \(x\) ≤ 10	2.05
learning_rate	Float	LogUniform	0.001 ≤ \(x\) ≤ 0.1	0.0381
reg_method	Categorical	Uniform	L2, AdamW	AdamW

Final Result

Searching selected the following configuration:

{
    'embedding_size': 182,
    'regularization': 2.045564699262321,
    'learning_rate': 0.03805132726783247,
    'reg_method': 'AdamW',
    'epochs': 4
}

With these metrics:

{
    'RBP': 0.02780012324782274,
    'LogRBP': 0.2115251448513078,
    'NDCG': 0.2129614179804314,
    'RecipRank': 0.10860804526319909,
    'RMSE': 0.9312385413697157,
    'TrainTask': 'c545f414-37fa-4e5c-862b-a22da158fc6d',
    'TrainTime': None,
    'TrainCPU': None,
    'max_epochs': 50,
    'done': False,
    'training_iteration': 4,
    'trial_id': '0d832_00022',
    'date': '2025-05-06_11-17-08',
    'timestamp': 1746544628,
    'time_this_iter_s': 0.7267158031463623,
    'time_total_s': 3.288360357284546,
    'pid': 475067,
    'hostname': 'gracehopper1',
    'node_ip': '192.168.225.60',
    'config': {
        'embedding_size': 182,
        'regularization': 2.045564699262321,
        'learning_rate': 0.03805132726783247,
        'reg_method': 'AdamW',
        'epochs': 4
    },
    'time_since_restore': 3.288360357284546,
    'iterations_since_restore': 4
}

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:

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?