PCLR Fold 3

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	5.21%
Test F1 score	0.0587
Hierarchical loss	0.75238147
P-adic loss (total)	3786.58087211
P-adic loss (mean)	0.81484417
Prime base	83
Training samples	18,450
Test samples	4,647
Trained at	2026-10-12T05:36:54+11:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	242	5.21%	0.000000	0.000000
p^4	10	0.22%	0.000000	0.000000
p^3	135	2.91%	0.000002	0.000236
p^2	348	7.49%	0.000145	0.050515
p^1	127	2.73%	0.012048	1.530120
p^0	3,785	81.45%	1.000000	3785.000000

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 83). Lower values indicate closer predictions in the taxonomy hierarchy. This metric is shared with the umllr model for comparison.