PCLR Fold 2

Back to PCLR overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	7.18%
Test F1 score	0.0644
Hierarchical loss	0.76060547
P-adic loss (total)	1215.15069617
P-adic loss (mean)	0.79369738
Prime base	79
Training samples	6,159
Test samples	1,531
Trained at	2026-07-14T05:31:19+10:00

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	110	7.18%	0.000000	0.000000
p^4	7	0.46%	0.000000	0.000000
p^3	39	2.55%	0.000002	0.000079
p^2	71	4.64%	0.000160	0.011376
p^1	90	5.88%	0.012658	1.139241
p^0	1,214	79.29%	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.