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 48
min_nbrs Integer Uniform 1 ≤ \(x\) ≤ 5 1
min_sim Float LogUniform 1e-06 ≤ \(x\) ≤ 0.1 9.74e-06

Final Result

Searching selected the following configuration:

{'max_nbrs': 48, 'min_nbrs': 1, 'min_sim': 9.742376138577668e-06}

With these metrics:

{
    'RBP': 0.07470347937379367,
    'LogRBP': 1.2000113598061093,
    'NDCG': 0.2992899265649201,
    'RecipRank': 0.27103613679466865,
    'TrainTask': 'f02e2201-f77d-46ec-ba6d-ba78cc09539b',
    'TrainTime': 0.15102143498370424,
    'TrainCPU': 0.14288600000000007,
    'TestTask': 'ded96963-c450-44cd-a307-42ccb4bdae75',
    'TestTime': 12.825546710984781,
    'TestCPU': 12.866875,
    'timestamp': 1746436910,
    'checkpoint_dir_name': None,
    'done': True,
    'training_iteration': 1,
    'trial_id': '30542_00070',
    'date': '2025-05-05_05-21-50',
    'time_this_iter_s': 13.64342975616455,
    'time_total_s': 13.64342975616455,
    'pid': 3390692,
    'hostname': 'CCI-ws21',
    'node_ip': '10.248.127.152',
    'config': {'max_nbrs': 48, 'min_nbrs': 1, 'min_sim': 9.742376138577668e-06},
    'time_since_restore': 13.64342975616455,
    'iterations_since_restore': 1,
    'experiment_tag': '70_max_nbrs=48,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.