| Parameter | Type | Distribution | Values | Selected | 
|---|---|---|---|---|
| embedding_size_exp | Integer | Uniform | 3 ≤ \(x\) ≤ 10 | 9 | 
| regularization.user | Float | LogUniform | 1e-05 ≤ \(x\) ≤ 1 | 0.127 | 
| regularization.item | Float | LogUniform | 1e-05 ≤ \(x\) ≤ 1 | 0.0319 | 
| damping.user | Float | LogUniform | 1e-12 ≤ \(x\) ≤ 100 | 8.25e-09 | 
| damping.item | Float | LogUniform | 1e-12 ≤ \(x\) ≤ 100 | 3.05 | 
ALS BiasedMF
This page analyzes the hyperparameter tuning results for biased matrix factorization with ALS.
Parameter Search Space
Final Result
Searching selected the following configuration:
{ 'embedding_size_exp': 9, 'regularization': {'user': 0.12720881694628106, 'item': 0.03193413440384729}, 'damping': {'user': 8.25385084467054e-09, 'item': 3.0462652993007437}, 'epochs': 4 }
With these metrics:
{ 'RBP': 0.015325934576731868, 'DCG': 0.6470205695226469, 'NDCG': 0.18166538154629294, 'RecipRank': 0.06452764049284829, 'Hit10': 0.14072847682119205, 'RMSE': 0.8106752038002014, 'max_epochs': 30, 'epoch_train_s': 11.145907836034894, 'epoch_measure_s': 3.00909839104861, 'done': False, 'training_iteration': 4, 'trial_id': 'db4403fb', 'date': '2025-09-30_15-03-03', 'timestamp': 1759258983, 'time_this_iter_s': 14.159525156021118, 'time_total_s': 55.93441438674927, 'pid': 250374, 'hostname': 'CCI-ws21', 'node_ip': '10.248.127.152', 'config': { 'embedding_size_exp': 9, 'regularization': {'user': 0.12720881694628106, 'item': 0.03193413440384729}, 'damping': {'user': 8.25385084467054e-09, 'item': 3.0462652993007437}, 'epochs': 4 }, 'time_since_restore': 55.93441438674927, 'iterations_since_restore': 4 }
Parameter Analysis
Embedding Size
The embedding size is the hyperparameter that most affects the model’s fundamental logic, so let’s look at performance as a fufnction of it:
Learning Parameters
Iteration Completion
How many iterations, on average, did we complete?
How did the metric progress in the best result?
How did the metric progress in the longest results?