PCNN Fold 1

Back to PCNN overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	15.04%
Test F1 score	0.1245
Hierarchical loss	0.78610617
P-adic loss (total)	1238.24910000
P-adic loss (mean)	0.68487230
Prime base	79
Hidden layer size	27
Max tags	32
Training samples	7,302
Test samples	1,808

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	272	15.04%	0.000000	0.000000
p^4	1	0.06%	0.000000	0.000000
p^3	69	3.82%	0.000002	0.000140
p^2	132	7.30%	0.000160	0.021150
p^1	97	5.37%	0.012658	1.227848
p^0	1,237	68.42%	1.000000	1237.000000

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.