PCLR Fold 4

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	5.99%
Test F1 score	0.0571
Hierarchical loss	0.75491691
P-adic loss (total)	2167.53188087
P-adic loss (mean)	0.79600877
Prime base	79
Training samples	10,836
Test samples	2,723
Trained at	2026-05-25T05:32:00+10:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	163	5.99%	0.000000	0.000000
p^4	1	0.04%	0.000000	0.000000
p^3	37	1.36%	0.000002	0.000075
p^2	159	5.84%	0.000160	0.025477
p^1	198	7.27%	0.012658	2.506329
p^0	2,165	79.51%	1.000000	2165.000000

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.