Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/akensert/molgraph
Graph neural networks for molecular machine learning. Implemented and compatible with TensorFlow and Keras.
https://github.com/akensert/molgraph
bioinformatics cheminformatics deep-learning graph-algorithms graph-neural-networks graphs keras molecular-graphs tensorflow
Last synced: 2 months ago
JSON representation
Graph neural networks for molecular machine learning. Implemented and compatible with TensorFlow and Keras.
- Host: GitHub
- URL: https://github.com/akensert/molgraph
- Owner: akensert
- License: mit
- Created: 2022-08-12T14:10:06.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-03-12T14:18:37.000Z (11 months ago)
- Last Synced: 2024-04-14T06:55:40.535Z (10 months ago)
- Topics: bioinformatics, cheminformatics, deep-learning, graph-algorithms, graph-neural-networks, graphs, keras, molecular-graphs, tensorflow
- Language: Python
- Homepage:
- Size: 1.29 MB
- Stars: 44
- Watchers: 3
- Forks: 5
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
Awesome Lists containing this project
README
**Graph Neural Networks** with **TensorFlow** and **Keras**. Focused on **Molecular Machine Learning**.
## Quick start
Benchmark the performance of MolGraph [here](https://github.com/akensert/molgraph/blob/main/examples/GNN-Benchmarking.ipynb), and implement a complete model pipeline with MolGraph [here](https://github.com/akensert/molgraph/blob/main/examples/QSAR-GNN-Tutorial.ipynb).
## Highlights
Build a Graph Neural Network with Keras' [Sequential](https://www.tensorflow.org/api_docs/python/tf/keras/Sequential) API:
```python
from molgraph import GraphTensor
from molgraph import layers
from tensorflow import keras
g = GraphTensor(node_feature=[[4.], [2.]], edge_src=[0], edge_dst=[1])
model = keras.Sequential([
layers.GNNInput(type_spec=g.spec),
layers.GATv2Conv(units=32),
layers.GATv2Conv(units=32),
layers.Readout(),
keras.layers.Dense(units=1),
])
pred = model(g)
# Save and load Keras model
model.save('/tmp/gatv2_model.keras')
loaded_model = keras.models.load_model('/tmp/gatv2_model.keras')
loaded_pred = loaded_model(g)
assert pred == loaded_pred
```
Combine outputs of GNN layers to improve predictive performance:
```python
model = keras.Sequential([
layers.GNNInput(type_spec=g.spec),
layers.GNN([
layers.FeatureProjection(units=32),
layers.GINConv(units=32),
layers.GINConv(units=32),
layers.GINConv(units=32),
]),
layers.Readout(),
keras.layers.Dense(units=128),
keras.layers.Dense(units=1),
])
model.summary()
```
## Installation
For **CPU** users:
pip install molgraph
For **GPU** users:
pip install molgraph[gpu]
## Implementations
- **Tensors**
- [Graph tensor](http://github.com/akensert/molgraph/tree/main/molgraph/tensors/graph_tensor.py)
- A composite tensor holding graph data; compatible with `tf.data.Dataset`, `keras.Sequential` and much more.
- **Layers**
- [Graph convolutional layers](http://github.com/akensert/molgraph/tree/main/molgraph/layers/convolutional/)
- [Graph attentional layers](http://github.com/akensert/molgraph/tree/main/molgraph/layers/attentional/)
- [Graph message passing layers](http://github.com/akensert/molgraph/tree/main/molgraph/layers/message_passing/)
- [Graph readout layers](http://github.com/akensert/molgraph/tree/main/molgraph/layers/readout/)
- [Preprocessing layers](https://github.com/akensert/molgraph/tree/main/molgraph/layers/preprocessing/)
- [Postprocessing layers](https://github.com/akensert/molgraph/tree/main/molgraph/layers/postprocessing/)
- [Positional encoding layers](https://github.com/akensert/molgraph/tree/main/molgraph/layers/positional_encoding)
- **Models**
- [Graph neural networks](https://github.com/akensert/molgraph/tree/main/molgraph/models/)
- [Saliency mapping](https://github.com/akensert/molgraph/tree/main/molgraph/models/interpretability/)
## Overview
## Documentation
See [readthedocs](https://molgraph.readthedocs.io/en/latest/)
## Papers
- [MolGraph: a Python package for the implementation of molecular graphs and graph neural networks with TensorFlow and Keras](https://doi.org/10.48550/arXiv.2208.09944)
- [A hands-on tutorial on quantitative structure-activity relationships using fully expressive graph neural networks](https://doi.org/10.1016/j.aca.2024.343046)