mfinzi/equivariant-MLP
A library for programmatically generating equivariant layers through constraint solving
This project helps machine learning practitioners build models that inherently understand and respect symmetries in their data, like rotations or permutations. You provide the type of symmetry and the format of your input and output data (e.g., vectors, tensors, or scalars), and it automatically generates the specialized model components. This is ideal for researchers and engineers working with structured scientific or physical data where underlying symmetries are crucial for accurate predictions.
281 stars. No commits in the last 6 months.
Use this if you need to build deep learning models for small to medium-sized datasets where physical or mathematical symmetries (like those in particle physics, molecular simulations, or graph data) are fundamental to the problem and you want to ensure your model respects these symmetries.
Not ideal if you are working with very large datasets like high-resolution images, large voxel grids, or extensive point clouds, as its performance will be limited to that of a standard Multilayer Perceptron.
Stars
281
Forks
26
Language
Jupyter Notebook
License
MIT
Category
Last pushed
May 08, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mfinzi/equivariant-MLP"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pyg-team/pytorch_geometric
Graph Neural Network Library for PyTorch
a-r-j/graphein
Protein Graph Library
raamana/graynet
Subject-wise networks from structural MRI, both vertex- and voxel-wise features (thickness, GM...
pykale/pykale
Knowledge-Aware machine LEarning (KALE): accessible machine learning from multiple sources for...
dmlc/dgl
Python package built to ease deep learning on graph, on top of existing DL frameworks.