PCNN Fold 2

Back to PCNN overview · Back to main index

Fold metrics

MetricValue
Test accuracy12.15%
Test F1 score0.0854
Hierarchical loss0.79585487
P-adic loss (total)1492.98930000
P-adic loss (mean)0.56001097
Prime base79
Hidden layer size27
Max tags32
Training samples10,606
Test samples2,666

P-adic loss breakdown

AgreementCountShareCost per mistakeTotal contribution
Exact match32412.15%0.0000000.000000
p^31385.18%0.0000020.000280
p^232612.23%0.0001600.052235
p^139014.63%0.0126584.936709
p^01,48855.81%1.0000001488.000000

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.