Unconstrained Logistic Regression

Model overview

Unconstrained logistic regression classifier using L1 (Lasso) regularization to predict taxonomy IDs from product tags. Unlike parameter-constrained models, this classifier uses ALL available tags and relies on L1 regularization to achieve sparsity by driving unimportant coefficients to zero.

5 CV folds

50.74% Mean accuracy

0.5496 Mean F1

0.21352673 Mean p-adic loss

3,146 Avg non-zero params

Total train samples: 24,724
Total test samples: 6,181
Number of tags (input features): 2,797
Avg non-zero parameters: 3,146 / 1,025,187 (99.7% sparse)
L1 regularization (C): 1.0000

Cross-validation results

Fold	Accuracy	F1	P-adic loss (mean)	Non-zero params	Details
0	51.80%	0.5612	0.20854515	3,070	View
1	50.69%	0.5509	0.20010882	3,231	View
2	51.61%	0.5639	0.20363934	3,065	View
3	49.24%	0.5273	0.23190402	3,119	View
4	50.36%	0.5445	0.22343632	3,247	View