PCLR Fold 0

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	7.06%
Test F1 score	0.0703
Hierarchical loss	0.76477160
P-adic loss (total)	1122.45416902
P-adic loss (mean)	0.73363018
Prime base	79
Training samples	6,160
Test samples	1,530
Trained at	2026-07-14T05:31:07+10:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	108	7.06%	0.000000	0.000000
p^4	3	0.20%	0.000000	0.000000
p^3	37	2.42%	0.000002	0.000075
p^2	148	9.67%	0.000160	0.023714
p^1	113	7.39%	0.012658	1.430380
p^0	1,121	73.27%	1.000000	1121.000000

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.