arnab39/equiadapt
Library to make any existing neural network architecture equivariant
This tool helps machine learning engineers and researchers adapt existing neural network models to better handle transformed data, such as rotated images or shifted point clouds. It takes your pre-trained model and inputs (like images or 3D point cloud data) and outputs predictions that are consistent regardless of how the input data is oriented or positioned, without requiring you to retrain your original model from scratch. This is ideal for those working with computer vision, robotics, or scientific modeling where data symmetries are crucial.
No commits in the last 6 months.
Use this if you need your existing neural networks, especially large pre-trained models, to produce consistent results even when input data is rotated, translated, or otherwise transformed.
Not ideal if you are building a neural network from the ground up and prefer to design equivariance directly into the architecture rather than adapting an existing one.
Stars
58
Forks
5
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Nov 03, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/arnab39/equiadapt"
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.