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

46.21% Mean accuracy

0.5058 Mean F1

0.25710750 Mean p-adic loss

6,672 Avg non-zero params

Total train samples: 54,236
Total test samples: 13,559
Number of tags (input features): 4,981
Avg non-zero parameters: 6,672 / 3,334,951 (99.8% sparse)
L1 regularization (C): 1.0000

Cross-validation results

Fold	Accuracy	F1	P-adic loss (mean)	Non-zero params	Details
0	45.11%	0.4937	0.27892279	6,588	View
1	46.77%	0.5104	0.26281342	6,750	View
2	46.56%	0.5092	0.27340997	6,620	View
3	46.15%	0.5077	0.26083189	6,677	View
4	46.46%	0.5079	0.20955944	6,725	View