Impeller: A Path-Based Heterogeneous Graph Learning Method for Spatial Transcriptomic Data Imputation
Clone the repository.
git clone https://github.com/aicb-ZhangLabs/Impeller.git
Make sure you have the required dependencies installed. Below are the packages and their specific versions required:
```bash
numpy==1.22.4
scipy==1.10.1
scikit-learn==1.2.1
torch==1.12.1
torch-geometric==2.2.0
torch-sparse==0.6.16
- data.py: Handles data preprocessing and dataset management. (The data can be accessed at https://github.com/LieberInstitute/spatialLIBD)
- model.py: Contains the model architecture for Impeller.
- train.py: Code for training the Impeller model.
- utils.py: Utility functions supporting data handling, model operations, etc.
python train.py
For questions about the data and code, please contact zihend1@uci.edu. We will do our best to provide support and address any issues. We appreciate your feedback!
You can find more information about the light version of Impeller here: https://pypi.org/project/impeller/0.1.2/