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 46
min_nbrs Integer Uniform 1 ≤ \(x\) ≤ 5 1
min_sim Float LogUniform 1e-06 ≤ \(x\) ≤ 0.1 4.66e-06

Final Result

Searching selected the following configuration:

{'max_nbrs': 46, 'min_nbrs': 1, 'min_sim': 4.659001835443387e-06}

With these metrics:

{
    'RBP': 0.0007900993022694542,
    'LogRBP': -3.349111951555653,
    'NDCG': 0.1712984244422581,
    'RecipRank': 0.013951692071563188,
    'RMSE': 0.9444416565081429,
    'TrainTask': '3f4ee206-8dca-45b9-9840-5eb31252545a',
    'TrainTime': 0.03322624001884833,
    'TrainCPU': 0.03645300000000057,
    'TestTask': '82adb772-58ea-4d23-82cc-2da97d88eb12',
    'TestTime': 0.8057771520107053,
    'TestCPU': 0.8185729999999996,
    'timestamp': 1746654848,
    'checkpoint_dir_name': None,
    'done': True,
    'training_iteration': 1,
    'trial_id': 'c7cb8_00020',
    'date': '2025-05-07_17-54-08',
    'time_this_iter_s': 1.446552038192749,
    'time_total_s': 1.446552038192749,
    'pid': 1451144,
    'hostname': 'gracehopper1',
    'node_ip': '192.168.225.60',
    'config': {'max_nbrs': 46, 'min_nbrs': 1, 'min_sim': 4.659001835443387e-06},
    'time_since_restore': 1.446552038192749,
    'iterations_since_restore': 1,
    'experiment_tag': '20_max_nbrs=46,min_nbrs=1,min_sim=0.0000'
}

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.