text-vectors

Text vector leverages GloVe using the provided by the Stanford Group, and other data sources within gensim-data.

The goal is to provide pre-trained models or sensible defaults for a variety of models that can interact with scikit-learn for the purposes of feature extraction. This is an opinionated approach to make it easier to get started with Machine Learning with text mining in a more automated approach.

Installation

pip install text_vectors

Example Usage

# by default uses GloVe 50d
from text_vectors import TextVec
tv = TextVec()

tv.fit_transform(["hello world".split(" ")])

# loading fasttext model with 300d
import gensim.downloader as api
model = api.load("fasttext-wiki-news-subwords-300")
model = TextVec(model)

model.transform([["hello world".split(" ")]])