Metadata-Version: 2.1
Name: danteai
Version: 1.1.1
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: python-dotenv
Requires-Dist: transformers
Requires-Dist: torch
Requires-Dist: tensorflow
Requires-Dist: datasets

# danteai

Danteai is a convenient Python package for training your own text-based AI model and using it in practice to generate responses based on specified parameters. The package is fully tailored for the tasks of Dante (an application for article generation). If you need to modify the default settings, pay attention to the `minions.py`, `create_or_train.py`, and `get_generated_article.py` files.

## Installation

Install the package using pip:

```bash
pip install danteai
```

## Example Usage

```bash
echo MODEL_STORE_DIR='./model' > .env
export $(cat .env | xargs)
```

```python
from danteai import (
    create_or_train,
    get_generated_article,
)
from dotenv import load_dotenv

load_dotenv()

"""
    data = [
        {
            'tags': [],
            'text_example': '',
            'short_news_description': '',
            'response': ''
        },
        ...
    ]
"""

create_or_train_response = create_or_train(
    data=[],
    create_new_model=True
)
print(create_or_train_response)
# if only training, using False for 'create_new_model'

generated_article = get_generated_article(
    tags=[],
    text_example='',
    short_news_description=''
)
print(generated_article)
```
