Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/blazerye/DrugAssist
DrugAssist: A Large Language Model for Molecule Optimization
https://github.com/blazerye/DrugAssist
ai-for-science drug-discovery instruction-datasets instruction-tuning large-language-models molecule-generation molecule-optimization
Last synced: 3 months ago
JSON representation
DrugAssist: A Large Language Model for Molecule Optimization
- Host: GitHub
- URL: https://github.com/blazerye/DrugAssist
- Owner: blazerye
- Created: 2023-12-27T07:48:51.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2024-07-26T03:27:41.000Z (4 months ago)
- Last Synced: 2024-07-27T04:01:59.767Z (4 months ago)
- Topics: ai-for-science, drug-discovery, instruction-datasets, instruction-tuning, large-language-models, molecule-generation, molecule-optimization
- Language: Python
- Homepage: https://arxiv.org/abs/2401.10334
- Size: 7.02 MB
- Stars: 126
- Watchers: 3
- Forks: 10
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- top-life-sciences - **blazerye/DrugAssist** - for-science`, `drug-discovery`, `instruction-datasets`, `instruction-tuning`, `large-language-models`, `molecule-generation`, `molecule-optimization`<br><img src='https://github.com/HubTou/topgh/blob/main/icons/gstars.png'> 123 <img src='https://github.com/HubTou/topgh/blob/main/icons/forks.png'> 10 <img src='https://github.com/HubTou/topgh/blob/main/icons/watchers.png'> 3 <img src='https://github.com/HubTou/topgh/blob/main/icons/code.png'> Python | (Ranked by starred repositories)
README
🐹 DrugAssist
A Large Language Model for Molecule Optimization
## 📌 Contents
- [Install](#install)
- [Dataset](#dataset)
- [Train](#train)
- [Demo](#demo)
- [About](#about)
## 🛠️ Install
1. Clone this repository and navigate to DrugAssist folder
```bash
git clone https://github.com/blazerye/DrugAssist.git
cd DrugAssist
```
2. Install Package
```Shell
conda create -n drugassist python=3.8 -y
conda activate drugassist
pip install -r requirements.txt
```
## 🤗 Dataset
We release the dataset on Hugging Face at [blazerye/MolOpt-Instructions](https://huggingface.co/datasets/blazerye/MolOpt-Instructions), and you can use it for training.
## 🚆 Train
You can use LoRA to finetune `Llama2-7B-Chat` model on the `MolOpt-Instructions` dataset, the running command is as follows:
```Shell
sh run_sft_lora.sh
```
## 👀 Demo
#### Step 1: Merge model weights
You can merge LoRA weights to generate full model weights using the following command:
```Shell
python merge_model.py \
--base_model $BASE_MODEL_PATH \
--lora_model $LORA_MODEL_PATH \
--output_dir $OUTPUT_DIR \
--output_type huggingface \
--verbose
```
Alternatively, you can download our DrugAssist model weights from [blazerye/DrugAssist-7B](https://huggingface.co/blazerye/DrugAssist-7B).
#### Step 2: Launch web demo
You can use gradio to launch web demo by running the following command:
```Shell
python gradio_service.py \
--base_model $FULL_MODEL_PATH \
--ip $IP \
--port $PORT
```
#### Deploy the Quantized Model and Use Text-Generation-WebUI For Inference
In order to deploy DrugAssist model on devices with lower hardware configurations (such as personal laptops without GPUs), we used [llama.cpp](https://github.com/ggerganov/llama.cpp) to perform 4-bit quantization on the [DrugAssist-7B](https://huggingface.co/blazerye/DrugAssist-7B) model, resulting in the [DrugAssist-7B-4bit](https://huggingface.co/blazerye/DrugAssist-7B/blob/main/DrugAssist-7B-4bit.gguf) model. You can use the text-generation-webui tool to load and use this quantized model. For specific methods, please refer to the [quantized_model_deploy.md](./quantized_model_deploy.md).
## ⚖️ Evaluate
After deploying the [DrugAssist-7B](https://huggingface.co/blazerye/DrugAssist-7B) model, you can refer to the [evaluate.md](./evaluate/evaluate.md) document and run the evaluate script to verify the molecular optimization results.
## 📝 About
### Citation
If you find DrugAssist useful for your research and applications, please cite using this BibTeX:
```bibtex
@article{ye2023drugassist,
title={DrugAssist: A Large Language Model for Molecule Optimization},
author={Ye, Geyan and Cai, Xibao and Lai, Houtim and Wang, Xing and Huang, Junhong and Wang, Longyue and Liu, Wei and Zeng, Xiangxiang},
journal={arXiv preprint arXiv:2401.10334},
year={2023}
}
```
### Acknowledgements
We appreciate [LLaMA](https://github.com/facebookresearch/llama), [Chinese-LLaMA-Alpaca-2](https://github.com/ymcui/Chinese-LLaMA-Alpaca-2), [Alpaca](https://crfm.stanford.edu/2023/03/13/alpaca.html), [iDrug](https://drug.ai.tencent.com) and many other related works for their open-source contributions.