Pinned Loading

1. thousand-language-morphology thousand-language-morphology Public

   Tools for building very wide (but shallow) language models. This scrapes bible.com for all languages and matches up verses with those from a parsed Greek New Testament -- to produce vocabulary and …

   Jupyter Notebook 5 1

2. padiclinear padiclinear Public

   Code for performing linear regression that minimises the sum of the p-adic distance to the residuals

   Jupyter Notebook 1 1

3. endless-summer endless-summer Public

   A repository of data-science style analyses of Sydney's weather

   Jupyter Notebook

4. leaftop leaftop Public

   Giant dataset of vocabulary extracted from low-resource languages.

   4 1

5. cross-embedding-clustering cross-embedding-clustering Public

   Studying how different embeddings cluster words

   Jupyter Notebook 1