Metadata-Version: 2.1
Name: antakia
Version: 0.2.8
Summary: AI made Xplainable
Keywords: XAI,explainability,AI,ML,machine learning,artificial intelligence
Author: Antoine EDY
Author-email: antoineedy@outlook.fr
Requires-Python: >=3.11,<3.12
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Requires-Dist: antakia-ac (>=0.2.4)
Requires-Dist: antakia-core
Requires-Dist: ipyvuetify (<1.9.0)
Requires-Dist: ipywidgets (>=8.1.1,<9.0.0)
Requires-Dist: lime (>=0.2.0.1,<0.3.0.0)
Requires-Dist: plotly (>=5.18.0,<6.0.0)
Requires-Dist: python-dotenv (>=1.0.0,<2.0.0)
Requires-Dist: requests (>=2.31.0,<3.0.0)
Requires-Dist: shap (>=0.43.0,<0.44.0)
Description-Content-Type: text/markdown

# Welcome to AntakIA !

AntakIA is an open-source tool from AI-vidence to explain ML black box models. See [doc](https://doc.antakia.ai).

Here is a quick overview on AntakIA:
![AntakIA demo](/docs/img/antakia.gif)

See full video on [Youtube](https://youtu.be/wQFC_20OIOM).

## Install

AntakIA is available on [PyPI](https://pypi.org/project/antakia/) :

```
pip install antakia
```

## Running example notebooks

Once you've installed `antakia`, you can download some of our notebook examples from our repo [here](https://github.com/AI-vidence/antakia/tree/main/examples).

Then, launch a Jupyter server from the notebook file (`.ipynb`) location :

```
jupyter notebook # or jupyter lab
```

You'll find a complete tutorial for our California housing example here : https://doc.antakia.ai

> [!IMPORTANT] 
If you're using a virtual env, it's handy to be able to chose it from Jupyter's kernel list. Do the following :

```
python -m ipykernel install --user --name your_venv_name --display-name "My great virtual env"
```

## Online demo

You can give `antakia` a try online  : https://demo.antakia.ai 

Log as `demo`  with password `antakia`.
Note it runs on a simple server and may be busy. You may want to log with other accounts : `demo1` to `demo5` are accepted with the same password.


## Run with Docker

> [!IMPORTANT] 
Be sure to have a Docker engine running on your computer (ie. launch Docker Desktop)

```
docker build -t antakia .
docker run -p 8888:8888 antakia
```

In your Terminal, click on the `http://127.0.0.1:8888/lab?token=WHATEVER_YOUR_TOKEN_IS URL` link.

