PCLR Fold 0

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	5.86%
Test F1 score	0.0597
Hierarchical loss	0.75941202
P-adic loss (total)	2042.66678703
P-adic loss (mean)	0.74795562
Prime base	79
Training samples	10,828
Test samples	2,731
Trained at	2026-05-25T05:31:21+10:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	160	5.86%	0.000000	0.000000
p^4	1	0.04%	0.000000	0.000000
p^3	33	1.21%	0.000002	0.000067
p^2	132	4.83%	0.000160	0.021150
p^1	367	13.44%	0.012658	4.645570
p^0	2,038	74.62%	1.000000	2038.000000

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.