PCLR Fold 1

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	5.68%
Test F1 score	0.0657
Hierarchical loss	0.74527779
P-adic loss (total)	4025.98338376
P-adic loss (mean)	0.86916740
Prime base	83
Training samples	18,465
Test samples	4,632
Trained at	2026-10-13T05:38:44+11:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	263	5.68%	0.000000	0.000000
p^4	4	0.09%	0.000000	0.000000
p^3	44	0.95%	0.000002	0.000077
p^2	217	4.68%	0.000145	0.031499
p^1	79	1.71%	0.012048	0.951807
p^0	4,025	86.90%	1.000000	4025.000000

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.