umllr P-adic Tag Regression

Using p-adic coefficients to predict taxonomy from tags

← Back to index

Overview

The umllr (Universal Machine Learning Linear Regression) model assigns p-adic integer coefficients to product tags and uses them to predict taxonomy encodings. Each taxonomy path is encoded as a p-adic integer (base 71), and tags are fitted to minimize p-adic distance on training data.

5 CV folds
0.3409 Average p-adic loss
43.97% Mean accuracy
0.4628 Mean F1
71 Prime base

Cross-validation results

FoldAccuracyF1P-adic loss (mean)Details
043.85%0.45630.35587078View details →
144.28%0.45510.33218043View details →
245.58%0.47440.31500944View details →
341.59%0.44740.34913012View details →
444.56%0.48090.35212978View details →