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 1
min_sim Float LogUniform 1e-06 ≤ \(x\) ≤ 0.1 0.0123

Final Result

Searching selected the following configuration:

{'max_nbrs': 49, 'min_nbrs': 1, 'min_sim': 0.012257043366936223}

With these metrics:

{
    'RBP': 0.20036164265392892,
    'NDCG': 0.403570584477069,
    'RecipRank': 0.36504210172817975,
    'TrainTask': '433c6c55-41a9-42c6-83f3-5d6ba01896f1',
    'TrainTime': 0.9203632419994392,
    'TrainCPU': 0.9195959999999999,
    'TestTask': '986d6714-f880-449e-afd0-2ab6d81488c5',
    'TestTime': 354.8347766950028,
    'TestCPU': 429.839769,
    'timestamp': 1743038348,
    'checkpoint_dir_name': None,
    'done': True,
    'training_iteration': 1,
    'trial_id': 'fa6f7_00021',
    'date': '2025-03-26_21-19-08',
    'time_this_iter_s': 359.0305850505829,
    'time_total_s': 359.0305850505829,
    'pid': 463868,
    'hostname': 'CCI-ws21',
    'node_ip': '10.248.127.152',
    'config': {'max_nbrs': 49, 'min_nbrs': 1, 'min_sim': 0.012257043366936223},
    'time_since_restore': 359.0305850505829,
    'iterations_since_restore': 1,
    'experiment_tag': '21_max_nbrs=49,min_nbrs=1,min_sim=0.0123'
}

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.