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 6.1e-05

Final Result

Searching selected the following configuration:

{'max_nbrs': 49, 'min_nbrs': 1, 'min_sim': 6.1013660177776704e-05}

With these metrics:

{
    'RBP': 0.19593546536868167,
    'NDCG': 0.35120666714974075,
    'RecipRank': 0.351096739043442,
    'TrainTask': '669a5bb3-bd1a-41ef-a8ac-ca199f817aa6',
    'TrainTime': 0.9614924420020543,
    'TrainCPU': 0.9558809999999998,
    'TestTask': '4a53912d-905b-4bda-a266-7b9f3d6f859a',
    'TestTime': 286.49350580399914,
    'TestCPU': 287.52697900000004,
    'timestamp': 1746229930,
    'checkpoint_dir_name': None,
    'done': True,
    'training_iteration': 1,
    'trial_id': '81a454aa',
    'date': '2025-05-02_19-52-10',
    'time_this_iter_s': 291.1766481399536,
    'time_total_s': 291.1766481399536,
    'pid': 1066160,
    'hostname': 'CCI-ws21',
    'node_ip': '10.248.127.152',
    'config': {'max_nbrs': 49, 'min_nbrs': 1, 'min_sim': 6.1013660177776704e-05},
    'time_since_restore': 291.1766481399536,
    'iterations_since_restore': 1,
    'experiment_tag': '59_max_nbrs=49,min_nbrs=1,min_sim=0.0001'
}

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.