PCLR Fold 4

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	7.18%
Test F1 score	0.0655
Hierarchical loss	0.75123444
P-adic loss (total)	1259.52817158
P-adic loss (mean)	0.83689580
Prime base	79
Training samples	6,185
Test samples	1,505
Trained at	2026-07-14T05:31:33+10:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	108	7.18%	0.000000	0.000000
p^4	15	1.00%	0.000000	0.000000
p^3	25	1.66%	0.000002	0.000051
p^2	57	3.79%	0.000160	0.009133
p^1	41	2.72%	0.012658	0.518987
p^0	1,259	83.65%	1.000000	1259.000000

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.