PCLR Fold 4

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	6.04%
Test F1 score	0.0567
Hierarchical loss	0.75485621
P-adic loss (total)	2073.07058065
P-adic loss (mean)	0.78704274
Prime base	79
Training samples	10,638
Test samples	2,634
Trained at	2026-05-20T05:31:28+10:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	159	6.04%	0.000000	0.000000
p^4	1	0.04%	0.000000	0.000000
p^3	39	1.48%	0.000002	0.000079
p^2	124	4.71%	0.000160	0.019869
p^1	241	9.15%	0.012658	3.050633
p^0	2,070	78.59%	1.000000	2070.000000

Tag Rank vs Coefficient Magnitude

Tag rank vs max coefficient magnitude — Scatter plot showing the relationship between tag battle ranking and maximum absolute coefficient value across all taxonomies. Shows which tags have the strongest influence in the model.

About p-adic loss

P-adic loss measures the distance between predicted and true taxonomy using p-adic metric (base 79). Lower values indicate closer predictions in the taxonomy hierarchy. This metric is shared with the umllr model for comparison.