NLP Embeddings Evaluation Tool

The NLP Embeddings Evaluation Tool is a command line tool to evaluate Natural Language Processing Embeddings using custom intrinsic and extrinsic tasks.

Installation

embedeval is available as pip package:

python -m pip install embedeval

NOTE: it might not be installable as of today using pip with PyPI. However, installing from source will work. Use . instead of embedeval in the pip command.

Getting started

Run the word-analogy Task on your Word Embedding:

embedeval embedding.vec -t word-analogy

Run the word-analogy and word-similarity Tasks on your Word Embedding:

embedeval embedding.vec -t word-analogy -t word-similarity

Documentation

The whole documentation of embedeval is available on Read The Docs.

Supported platforms

embedeval is supported on Windows, Mac and Linux

Contribution

Yes, we are looking for some contributors and people who spread out a word about embedeval. Help us to improve these piece of software. You don't know what to do? Just have a look at the Issues or create a new one. Please have a look at the Contributing Guidelines, too.

Project Information

embedeval is released under the MIT license, its documentation lives at Read The Docs, the code on GitHub, and the latest release on PyPI. It’s rigorously tested on Python 3.5+.

If you'd like to contribute to embedeval you're most welcome and we've written a little guide to get you started!

This project is published under MIT.
A Timo Furrer project.
- :tada: -