https://github.com/ur-whitelab/chemcrow-public

Chemcrow
https://github.com/ur-whitelab/chemcrow-public

Last synced: 11 days ago
JSON representation

Chemcrow

• Host: GitHub
• URL: https://github.com/ur-whitelab/chemcrow-public
• Owner: ur-whitelab
• License: mit
• Created: 2023-06-04T15:59:05.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-04-03T19:49:19.000Z (7 months ago)
• Last Synced: 2024-10-01T10:18:53.889Z (about 1 month ago)
• Language: Python
• Homepage:
• Size: 5.77 MB
• Stars: 593
• Watchers: 18
• Forks: 86
• Open Issues: 8
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-ai-agents - GitHub
• best-of-atomistic-machine-learning - GitHub - 36% open · ⏱️ 27.03.2024): (Language Models)

README

        
[![tests](https://github.com/ur-whitelab/chemcrow-public/actions/workflows/tests.yml/badge.svg)](https://github.com/ur-whitelab/chemcrow-public)

[![PyPI](https://img.shields.io/pypi/v/chemcrow)](https://img.shields.io/pypi/v/chemcrow)

[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/chemcrow)](https://img.shields.io/pypi/pyversions/chemcrow)

[![DOI:10.1101/2020.07.15.204701](https://zenodo.org/badge/DOI/10.48550/arXiv.2304.05376.svg)](https://doi.org/10.48550/arXiv.2304.05376)

[![DOI](https://zenodo.org/badge/649361700.svg)](https://zenodo.org/doi/10.5281/zenodo.10884638)

  

  

  




ChemCrow is an open source package for the accurate solution of reasoning-intensive chemical tasks.

Built with Langchain, it uses a collection of chemical tools including RDKit, paper-qa, as well as some relevant databases in chemistry, like Pubchem and chem-space.

## 🤗 Try it out in [HuggingFace](https://huggingface.co/spaces/doncamilom/ChemCrow)!

[![ChemCrow Demo](assets/hf-demo.png)](https://huggingface.co/spaces/doncamilom/ChemCrow)

## ⚠️ Note

This package does not contain all the tools described in the [ChemCrow paper](https://arxiv.org/abs/2304.05376) because

of API usage restrictions. This repo will not give the same results as that paper.

All the experiments have been released under [ChemCrow runs](https://github.com/ur-whitelab/chemcrow-runs).

## 👩‍💻 Installation

```

pip install chemcrow

```

## 🔥 Usage

First set up your API keys in your environment.

```

export OPENAI_API_KEY=your-openai-api-key

```

You can optionally use Serp API:

```

export SERP_API_KEY=your-serpapi-api-key

```

In a Python session:

```python

from chemcrow.agents import ChemCrow

chem_model = ChemCrow(model="gpt-4-0613", temp=0.1, streaming=False)

chem_model.run("What is the molecular weight of tylenol?")

```

## ✅ Citation

Bran, Andres M., et al. "ChemCrow: Augmenting large-language models with chemistry tools." arXiv preprint arXiv:2304.05376 (2023).

```bibtex

@article{bran2023chemcrow,

      title={ChemCrow: Augmenting large-language models with chemistry tools},

      author={Andres M Bran and Sam Cox and Oliver Schilter and Carlo Baldassari and Andrew D White and Philippe Schwaller},

      year={2023},

      eprint={2304.05376},

      archivePrefix={arXiv},

      primaryClass={physics.chem-ph},

      publisher={arXiv}

}

```