PCNN Fold 4

Back to PCNN overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	11.46%
Test F1 score	0.0872
Hierarchical loss	0.79275930
P-adic loss (total)	1506.07650000
P-adic loss (mean)	0.55309457
Prime base	79
Hidden layer size	27
Max tags	32
Training samples	10,836
Test samples	2,723

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	312	11.46%	0.000000	0.000000
p^3	117	4.30%	0.000002	0.000237
p^2	318	11.68%	0.000160	0.050953
p^1	476	17.48%	0.012658	6.025316
p^0	1,500	55.09%	1.000000	1500.000000

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.