PCLR Fold 3

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	5.33%
Test F1 score	0.0497
Hierarchical loss	0.74991041
P-adic loss (total)	1216.00839292
P-adic loss (mean)	0.79012891
Prime base	79
Training samples	6,151
Test samples	1,539
Trained at	2026-07-14T05:31:26+10:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	82	5.33%	0.000000	0.000000
p^4	3	0.19%	0.000000	0.000000
p^3	30	1.95%	0.000002	0.000061
p^2	52	3.38%	0.000160	0.008332
p^1	158	10.27%	0.012658	2.000000
p^0	1,214	78.88%	1.000000	1214.000000

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.