Metadata-Version: 2.3
Name: graphbook_huggingface
Version: 0.0.6
Summary: Graphbook Hugging Face Plugin for no-code Hugging Face AI pipelines
License: MIT
Keywords: huggingface,ml,workflow,pipelines,pytorch,data science,machine learning,ai
Author: Richard Franklin
Author-email: rsamf@graphbook.ai
Requires-Python: >=3.9,<4.0
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Project-URL: Documentation, https://docs.graphbook.ai
Project-URL: Homepage, https://graphbook.ai
Project-URL: Repository, https://github.com/graphbookai/graphbook
Description-Content-Type: text/markdown

<p align="center">
  <a href="https://graphbook.ai">
    <img src="assets/graphbook-hf-banner.png" alt="Logo" width=512>
  </a>

  <h1 align="center">Graphbook Hugging Face</h1>

  <p align="center">
    Build No Code Hugging Face AI Pipelines
  </p>
</p>

You can build efficient DAG workflows or AI pipelines without any code. This is a Graphbook plugin that lets you drag and drop Hugging Face models and datasets onto Graphbook workflows. This plugin contains a web panel for searching and drag-and-dropping models and datasets from [Huggingface Hub](https://huggingface.co/) onto their graphbook workflows.

<img src="assets/example-hf-pipeline.png" alt="Example Pipeline with Hugging Face" with=1024>

## Packaged Nodes

Graphbook Hugging Face contains the following nodes:

* `TransformersPipeline` step for model usage from transformers package
* `HuggingfaceDataset` step for dataset usage from the datasets package
* And numerous `Post Processing/*` steps for post processing of model outputs

## Getting started
1. `pip install graphbook_huggingface graphbook transformers datasets`
1. `graphbook --config hf.config.yaml`


