UKNN Explicit

This page analyzes the hyperparameter tuning results for the ItemKNN scorer in explicit-feedback mode.

Parameter Search Space

Parameter Type Distribution Values Selected
max_nbrs Integer Uniform 2 ≤ \(x\) ≤ 50 48
min_nbrs Integer Uniform 1 ≤ \(x\) ≤ 5 1
min_sim Float LogUniform 1e-06 ≤ \(x\) ≤ 0.1 0.0204

Final Result

Searching selected the following configuration:

{'max_nbrs': 48, 'min_nbrs': 1, 'min_sim': 0.020380060202381322}

With these metrics:

{
    'RBP': 0.013572146100776995,
    'NDCG': 0.3093858863448935,
    'RecipRank': 0.04327818388134123,
    'RMSE': 0.7784032179668486,
    'TrainTask': '8e8304bb-5250-490f-9b51-ab8f886498da',
    'TrainTime': 2.8933355978224427,
    'TrainCPU': 3.8924010000000004,
    'TestTask': '03619c54-825a-4bec-9a21-161e47c9d7a8',
    'TestTime': 1327.541703600902,
    'TestCPU': 3789.320911,
    'timestamp': 1742757072,
    'checkpoint_dir_name': None,
    'done': True,
    'training_iteration': 1,
    'trial_id': '747e4_00014',
    'date': '2025-03-23_19-11-12',
    'time_this_iter_s': 1335.8153076171875,
    'time_total_s': 1335.8153076171875,
    'pid': 3396354,
    'hostname': 'cci-p102',
    'node_ip': '10.246.250.62',
    'config': {'max_nbrs': 48, 'min_nbrs': 1, 'min_sim': 0.020380060202381322},
    'time_since_restore': 1335.8153076171875,
    'iterations_since_restore': 1,
    'experiment_tag': '14_max_nbrs=48,min_nbrs=1,min_sim=0.0204'
}

Parameter Behavior

Neighborhood Size

The neighborhood size is key argument to consider, let’s look at RMSE with repspect to that:

Filtering

There are two filtering parameters we also test — minimum neighbors and mininmum simialrity.