Molecular property prediction based on Graph Convolution Network published by Deep4Chem
Project description
D4C molecular property prediction
This project is a deep learning application designed to predict molecular properties. The models implemented in this project feature interpretable and hierarchical architectures, including conventional graph convolutional models.
How to start
- Place the CSV file for training in the "_Data" folder.
- The SMILES strings of molecules should be in the "compound" column.
- There needs to be at least one molecular property for each corresponding molecule.
- Check and choose the ID of the deep learning model in "network_refer.yaml".
- Run the "main.py" file as
python main.py -n [network_id] -d [data_file_name] -t [property_column]
For example,
python main.py -n GCN -d Aqsoldb -t Solubility
The graph cache file will be saved in "_Graph" folder. The trained result will be saved in "_Model" folder.
Code Attribution and Licensing Information
This project includes code from the GC-GNN by Adem Rosenkvist Nielsen Aouichaoui (arnaou@kt.dtu.dk), licensed under the MIT License. https://github.com/gsi-lab/GC-GNN/tree/main
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
D4CMPP-0.22.tar.gz
(32.6 kB
view hashes)
Built Distribution
D4CMPP-0.22-py3-none-any.whl
(44.1 kB
view hashes)