PCLR Fold 4

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	8.94%
Test F1 score	0.0845
Hierarchical loss	0.91722218
P-adic loss (total)	937.93578591
P-adic loss (mean)	0.75579032
Prime base	71
Training samples	4,940
Test samples	1,241
Trained at	2026-02-02T05:30:39+11:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	111	8.94%	0.000000	0.000000
p^4	5	0.40%	0.000000	0.000000
p^3	21	1.69%	0.000003	0.000059
p^2	102	8.22%	0.000198	0.020234
p^1	65	5.24%	0.014085	0.915493
p^0	937	75.50%	1.000000	937.000000

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