Mysteries Department
Sign inRequest access
MethodsNLP (cross-lingual embedding alignment)A <-> B alignment of distributional spaces (orthogonal Procrustes)
Method · NLP (cross-lingual embedding alignment)

A <-> B alignment of distributional spaces (orthogonal Procrustes)

Nlp (cross-lingual embedding alignment)Version 1.0Licence MIT10 runsUsed on 1 mystery

How it works

Compared with: Latin of the manuscript's genres, slot-grammar text and the manuscript shuffled across pages, all in the same lines.

Code Python

Download module (.py)methods_ind.pyView source

Runs and results 10 runs · 6 shown

RunMysteryMeasuredResultAgainstDate
RUN-238VoynichRandom edit10.011Latin 0.0 · slot 0.0 · shuffled 0.022 Oct
RUN-237VoynichA only mapped to edit10.011Latin 0.0 · slot 0.0 · shuffled 0.022 Oct
RUN-236VoynichA only words93.0Latin 64.0 · slot 10.0 · shuffled 12.022 Oct
RUN-235VoynichHeldout self alignment0.0Latin 0.03 · slot 0.0 · shuffled 0.022 Oct
RUN-234VoynichShared177.0Latin 219.0 · slot 265.0 · shuffled 278.022 Oct
RUN-155VoynichRandom edit10.018Latin 0.0 · slot 0.0 · shuffled 0.05922 Oct

Applied to

Voynich Manuscript
3 datasetsExploration split, Latin genre texts, ZL3b transcription
Private beta Downloads are for beta members

We're in a private beta while we get things right. Everything here is free to read; downloading datasets, sources and code is open to members for now.

Apply to the betaSign in
It takes a couple of minutes. We'll reply by email.