PCNN Fold 4

Back to PCNN overview · Back to main index

Fold metrics

Metric	Value
Test accuracy	16.87%
Test F1 score	0.1196
Hierarchical loss	0.77466230
P-adic loss (total)	703.22920000
P-adic loss (mean)	0.78049850
Prime base	79
Hidden layer size	27
Max tags	32
Training samples	3,679
Test samples	901

P-adic loss breakdown

Agreement	Count	Share	Cost per mistake	Total contribution
Exact match	152	16.87%	0.000000	0.000000
p^4	4	0.44%	0.000000	0.000000
p^3	16	1.78%	0.000002	0.000032
p^2	8	0.89%	0.000160	0.001282
p^1	18	2.00%	0.012658	0.227848
p^0	703	78.02%	1.000000	703.000000

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.