PCLR Fold 4

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	8.71%
Test F1 score	0.0882
Hierarchical loss	0.76703711
P-adic loss (total)	1306.59195520
P-adic loss (mean)	0.71126399
Prime base	79
Training samples	7,273
Test samples	1,837
Trained at	2026-03-20T05:30:38+11:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	160	8.71%	0.000000	0.000000
p^3	31	1.69%	0.000002	0.000063
p^2	139	7.57%	0.000160	0.022272
p^1	203	11.05%	0.012658	2.569620
p^0	1,304	70.99%	1.000000	1304.000000

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.