ULR Fold 3

Back to ULR overview · Back to main index

Fold metrics

MetricValue
Test accuracy50.29%
Test F1 score0.5313
Hierarchical loss0.95480896
P-adic loss (total)155.47242790
P-adic loss (mean)0.22532236
Prime base71
Number of tags (input features)1,640
Non-zero parameters1,551 / 359,379 (99.6% sparse)
L1 regularization (C)1.0000
Training samples2,733
Test samples690

P-adic loss breakdown

AgreementCountShareCost per mistakeTotal contribution
Exact match34750.29%0.0000000.000000
p^610.14%0.0000000.000000
p^500.00%0.0000000.000000
p^4101.45%0.0000000.000000
p^3365.22%0.0000030.000101
p^210915.80%0.0001980.021623
p^1324.64%0.0140850.450704
p^015522.46%1.000000155.000000

About L1 regularization

L1 (Lasso) regularization promotes sparsity by driving many coefficients to exactly zero. This model uses ALL available tags (1,640) but L1 regularization selects which features are actually used. The number of non-zero parameters (1,551) indicates how many coefficients the model actually uses.