Metadata-Version: 2.1
Name: nlpflow
Version: 0.0.2
Summary: A powerful toolkit for automating NLP workflows.
Home-page: https://github.com/GiannopoulosK/NLPFlow
Author: Konstantinos Giannopoulos
Author-email: giannopoulosk.data@gmail.com
License: Apache License 2.0
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Development Status :: 2 - Pre-Alpha
Requires-Python: >=3.7
Description-Content-Type: text/markdown
License-File: LICENSE

# NLPFlow

![Version](https://img.shields.io/badge/version-0.0.2-brightgreen) 
![License](https://img.shields.io/badge/license-Apache%202.0-blue)
![Python Version](https://img.shields.io/badge/python-3.7%2B-blue)

NLPFlow is a Python library designed to simplify the process of building and deploying natural language processing (NLP) applications. It leverages well-established tools like `spaCy`, `scikit-learn`, and `optuna` for text preprocessing and model development. NLPFlow provides a powerful toolkit for automating NLP workflows, making it easier for data scientists and developers to efficiently build models.

## Features

Although NLPFlow is in early-stage development, it already offers the following:

- A full text preprocessing pipeline with optional preprocessing steps.
- Development of Naive Bayes models with hyperparameter tuning using Bayesian optimization.

Planned features include:

- Tuning based on metrics beyond simple accuracy.
- Sentiment analysis using pretrained models.
- Named Entity Recognition (NER) using `spaCy`, with added functionality.
- Support for additional model types with hyperparameter tuning.
- Integration with popular tools such as MLFlow, GitHub Actions, and Docker.

## Installation

You can install NLPFlow using pip:

```bash
pip install nlpflow
```

## Documentation
Documentation is still under development. For example usage of the library, check the [examples.ipynb](https://github.com/GiannopoulosK/NLPFlow/blob/main/examples.ipynb) file.

## Contributing
For contributions, please contact me directly on LinkedIn [here](https://www.linkedin.com/in/kongiannopoulos/).

## License
This project is licensed under the [Apache License 2.0](https://github.com/GiannopoulosK/NLPFlow/blob/main/LICENSE).

## Acknowledgements
Thanks to the developers and contributors of the libraries this tool depends on! Let's continue improving the open-source community together!
