Metadata-Version: 2.1
Name: full
Version: 0.0.3.1
Project-URL: Documentation, https://github.com/maximedb/full
Project-URL: Issues, https://github.com/maximedb/full
Project-URL: Source, https://github.com/maximedb/full
Author-email: maximedb <maxime.debruyn@uantwerpen.be>
License-File: LICENSE.txt
Classifier: Development Status :: 4 - Beta
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: Implementation :: CPython
Classifier: Programming Language :: Python :: Implementation :: PyPy
Requires-Python: >=3.7
Requires-Dist: torch
Requires-Dist: transformers
Description-Content-Type: text/markdown

# FULL

Unsupervised evaluation of open-domain conversations using follow-ups likelihood.

-----

## Installation

```console
pip install full
```

## Example
We provide an example script using FULL which reproduces the results of the paper.

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/maximedb/full/blob/master/examples/results_reproduction.ipynb)

## Turn evaluation

```python
from full import FULL
eval_model = FULL()
conversation = ["Hi", "What's your name"]
response = "None of your business"
evaluation = eval_model.evaluate_turn(convesation, response)
print(evaluation)
```

## Conversation evaluation

```python
from full import FULL
eval_model = FULL()
conversation = ["Hi", "What's your name", "None of your business"]
evaluation = eval_model.evaluate_turn(convesation)
print(evaluation)
```

## License

`full` is distributed under the terms of the [MIT](https://spdx.org/licenses/MIT.html) license.
