GoldNLP is Golden's open-source NLP/NLU package and framework for running and creating AI/ML-based models and tools that can support ML practitioners and developers contributing to Golden's protocol community.
This repository will also include API services and end-to-end ML training/inference framworks for tools like search/disambiguation, named entity recognition, texts classification, relationship extraction, and more. GoldNLP is built with Spacy, Transformers, FastNN, and FastAPI.
Our models will also be hosted in in our golden-models repository and Hugging Face's Model Hub for pulling and installation.
GoldNLP requires Python 3.7+
pip install goldnlp
First we'll need to load our models so run the following to install some of Golden's and Spacy models:
pip install https://github.com/explosion/spacy-models/releases/download/en_core_web_md-3.1.0/en_core_web_md-3.1.0-py3-none-any.whl
pip install https://github.com/goldenrecursion/golden-models/releases/download/en_funding_round_model-0.0.1/en_funding_round_model-0.0.1-py3-none-any.whl
pip install https://github.com/goldenrecursion/golden-models/releases/download/en_ceo_model-0.0.1/en_ceo_model-0.0.1-py3-none-any.whl
Now we can get started with running inference on some of Golden's custom trained named-entity recognition models.
from goldnlp import create_document
text = "Tim Cook is the CEO of Apple."
doc = create_document(text=text, model="default")
entities = [(spn.text, spn.start_char, spn.end_char, spn.label_) for spn in doc.ents]
print(entities)Run docker-compose build and then docker-compose up.
Go to localhost:8888 to run our tutorials in jupyter lab.
For all things related to goldnlp and development, please contact the maintainer Andrew Chang at andrew@golden.co or @achang1618 for any quesions or comments.
For all other support, please reach out to support@golden.co.
Follow @golden to keep up with additional news!
This project is licensed under the terms of the Apache 2.0 license