kimhc6028/relational-networks

Pytorch implementation of "A simple neural network module for relational reasoning" (Relational Networks)

51
/ 100
Established

This tool helps researchers in artificial intelligence and machine learning evaluate how well their models can understand relationships between objects in images. It takes a dataset of simplified images containing various shapes and colors, along with corresponding questions about those objects and their interactions. The output indicates how accurately the model answers both simple object-recognition questions and complex relational questions, helping to benchmark its 'reasoning' capabilities.

818 stars. No commits in the last 6 months.

Use this if you are developing or evaluating AI models and need a benchmark to test their ability to understand spatial and property relationships between multiple objects in a visual scene.

Not ideal if you are looking for a tool to process real-world complex images or to perform general image classification or object detection tasks.

artificial-intelligence-research visual-reasoning machine-learning-benchmarking computer-vision-evaluation cognitive-ai
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 25 / 25

How are scores calculated?

Stars

818

Forks

160

Language

Python

License

BSD-3-Clause

Last pushed

Dec 06, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/kimhc6028/relational-networks"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.