Metadata-Version: 2.1
Name: green-agent
Version: 0.3.0
Summary: 
Author: Andrei Anton
Author-email: io@neuronq.ro
Requires-Python: >=3.9
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Requires-Dist: google-cloud-aiplatform (>=1.34.0,<2.0.0)
Requires-Dist: markdown (>=3.5,<4.0)
Requires-Dist: openai (>=0.28.1,<0.29.0)
Requires-Dist: python-dotenv (>=1.0.0,<2.0.0)
Description-Content-Type: text/markdown

# Green Agent

Library providing uniform (mostly) abstration across foundation-models APIs + nice "interface" helpsers for running experiments or crafting your own ChatGPT-style UI.

For now it supports chat-tuned LLMs from OpenAI and Google (Vertex AI) - more stuff to come.

## Usage

Inside a jupyter notebook:

```py
import green_agent as ga

ag1 = ga.AgentNbUI(ga.OpenAIAgent(
    "You are a sarcastic but helpful assistant.",
    model="gpt-3.5-turbo"
))
ag1.asks("Which foods would you suggest trying when visiting Lyon, France?")
```

## How to develop

First time:

```sh
poetry install --with test dev
poetry shell
python -m ipykernel install --user --name green-agent
```

In general:

```sh
poetry shell
```

Notebooks:

```sh
jupyter notebook  # classic UI
jupyter lab  # modern UI
```

### Managing dependencies with Poetry

```
poetry add <my-dependency>
poetry add <my-dependency> --group dev
poetry add <my-dependency> --group test
```

```
poetry export -o all-requirements.txt
```

