PCNN Fold 1

Back to PCNN overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	13.28%
Test F1 score	0.0968
Hierarchical loss	0.79795030
P-adic loss (total)	1454.45170000
P-adic loss (mean)	0.55176467
Prime base	79
Hidden layer size	27
Max tags	32
Training samples	10,636
Test samples	2,636

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	350	13.28%	0.000000	0.000000
p^3	121	4.59%	0.000002	0.000245
p^2	289	10.96%	0.000160	0.046307
p^1	427	16.20%	0.012658	5.405063
p^0	1,449	54.97%	1.000000	1449.000000

Tag Rank vs First-Layer Weight Magnitude

Tag rank vs max first-layer weight magnitude — Scatter plot showing the relationship between tag battle ranking and maximum absolute first-layer weight value across all hidden units. Shows which input features contribute most to the parameter constrained neural network's hidden representations.

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.