PCLR Fold 3

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	12.03%
Test F1 score	0.1138
Hierarchical loss	0.92002635
P-adic loss (total)	318.46919246
P-adic loss (mean)	0.46154955
Prime base	71
Training samples	2,733
Test samples	690
Trained at	2025-12-16T22:25:45+11:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	83	12.03%	0.000000	0.000000
p^4	10	1.45%	0.000000	0.000000
p^3	14	2.03%	0.000003	0.000039
p^2	93	13.48%	0.000198	0.018449
p^1	174	25.22%	0.014085	2.450704
p^0	316	45.80%	1.000000	316.000000

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.