PCNN Fold 2

Back to PCNN overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	11.55%
Test F1 score	0.0810
Hierarchical loss	0.79697780
P-adic loss (total)	1498.42420000
P-adic loss (mean)	0.55456114
Prime base	79
Hidden layer size	27
Max tags	32
Training samples	10,857
Test samples	2,702

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	312	11.55%	0.000000	0.000000
p^3	118	4.37%	0.000002	0.000239
p^2	355	13.14%	0.000160	0.056882
p^1	424	15.69%	0.012658	5.367089
p^0	1,493	55.26%	1.000000	1493.000000

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.