---
title: Flare
emoji: 🔥 
colorFrom: green
colorTo: blue
sdk: streamlit
pinned: false
python_version: 3.11.7
---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

# 🔥 FLARE
Fine-grained Learning for Aligment of spectra-molecule REpresentation

###  Yan Zhou Chen, Soha Hassoun
#### Department of Computer Science, Tufts University
This repository provides the implementation of FALRE. Given a mass spectrum, FLARE can be used to rank a set of molecular candidates. Further, FLARE enables visualization of peak-to-node attribution.

## 🛠 Set up
### Clone repository
```
git clone https://github.com/hassounlab/flare
cd flare
```
### Set up environment and install dependencies
```
conda create -n flare python=3.11
conda activate flare
pip install -r requirements.txt
```

## Usage
Modify params.yaml as necessary

```
# preprocess data
python subformula_assign/assign_subformulae.py --spec-files ../data/sample/data.tsv --output-dir ../data/sample/subformulae --labels-file ../data/sample/data.tsv --max-formulae 60

# train 
python train.py

# test
python test.py
```
## 🙏 Acknowledgments
- **Loss Objective**: [FILIP](https://arxiv.org/abs/2111.07783)
- **Training Data**: [MassSpecGym](https://github.com/pluskal-lab/MassSpecGym)
- **Subformula Assigner Code**: [MIST](https://github.com/samgoldman97/mist/tree/main_v2)