Uzbek 65 million web corpus lemmatized FastText model

FastText lemmatized model developed from 65 million Uzbek web corpus (10.5281/zenodo.19462612). The model can be used for efficient text representation via Word embeddings (vector representations). A simple Python script for loading and using the model. FastText is a lightweight, open-source library developed by Facebook AI Research for efficient text representation and classification. The FastText implementation used to produce the model, which can later be used for the model, was Gensim. The original corpus was lemmatized using UzbekLemma lemmatizer: https://pypi.org/project/UzbekLemma/.

Перевод пока недоступен