Metadata-Version: 2.1
Name: impeller
Version: 0.1.0
Summary: Impeller is a package for spatial transcriptomics imputation using path-based graph neural networks.
Home-page: https://github.com/aicb-ZhangLabs/Impeller
Author: Ziheng Duan
Author-email: duanziheng1206@gmail.com
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: torch
Requires-Dist: gdown
Requires-Dist: scanpy
Requires-Dist: pandas
Requires-Dist: numpy
Requires-Dist: scikit-learn
Requires-Dist: dgl

# Impeller

Impeller is a package for spatial transcriptomics imputation using path-based graph neural networks.

## Installation

You can install the package using pip:

```bash
pip install Impeller
```

## Usage
### Download Example Data
from Impeller import download_example_data
download_example_data()

### Load and Process Data
from Impeller import load_and_process_example_data
data, val_mask, test_mask, x, original_x = load_and_process_example_data()

### Train Model
from Impeller import create_args, train
args = create_args()
test_l1_distance, test_cosine_sim, test_rmse = train(args, data, val_mask, test_mask, x, original_x)
