IKNN Implicit

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

Parameter Search Space

Parameter Type Distribution Values Selected
max_nbrs Integer Uniform 2 ≤ \(x\) ≤ 50 49
min_nbrs Integer Uniform 1 ≤ \(x\) ≤ 5 2
min_sim Float LogUniform 1e-06 ≤ \(x\) ≤ 0.1 0.00311

Final Result

Searching selected the following configuration:

{'max_nbrs': 49, 'min_nbrs': 2, 'min_sim': 0.0031071803184892018}

With these metrics:

{
    'RBP': 0.150756389945179,
    'LogRBP': 1.9021499131946342,
    'NDCG': 0.36111877018395294,
    'RecipRank': 0.29916232280893124,
    'TrainTask': '224689cc-373e-4be4-8d00-5b898f5667a3',
    'TrainTime': 4.417997696029488,
    'TrainCPU': 4.492551,
    'TestTask': '55e8cac4-9d38-4efe-b759-54f189fad36f',
    'TestTime': 401.6771317179664,
    'TestCPU': 2455.3759990000003,
    'timestamp': 1746676493,
    'checkpoint_dir_name': None,
    'done': True,
    'training_iteration': 1,
    'trial_id': '764bb7f1',
    'date': '2025-05-07_23-54-53',
    'time_this_iter_s': 424.4349858760834,
    'time_total_s': 424.4349858760834,
    'pid': 1613086,
    'hostname': 'gracehopper1',
    'node_ip': '192.168.225.60',
    'config': {'max_nbrs': 49, 'min_nbrs': 2, 'min_sim': 0.0031071803184892018},
    'time_since_restore': 424.4349858760834,
    'iterations_since_restore': 1,
    'experiment_tag': '28_max_nbrs=49,min_nbrs=2,min_sim=0.0031'
}

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.