relational-networks and local-relational-nets
These are ecosystem siblings—both are independent PyTorch implementations of different relational reasoning architectures (relational networks vs. local relational networks) that can be studied or adapted separately but share a common conceptual foundation in how neural networks learn to reason about relationships between entities or spatial regions.
About relational-networks
kimhc6028/relational-networks
Pytorch implementation of "A simple neural network module for relational reasoning" (Relational Networks)
Implements relational reasoning through a specialized neural module that compares object pairs using concatenated feature representations, evaluated on the Sort-of-CLEVR visual question-answering task with both binary and ternary relation types. The architecture combines a CNN feature extractor with a relation module that processes pairwise object interactions, achieving 89% accuracy on relational questions compared to 66% for standard CNN+MLP baselines. Supports PyTorch training with configurable relation types and includes dataset generation utilities for the synthetic CLEVR benchmark.
About local-relational-nets
gan3sh500/local-relational-nets
A Pytorch implementation for the paper Local Relational Networks for Image Recognition (https://arxiv.org/pdf/1904.11491.pdf)
Scores updated daily from GitHub, PyPI, and npm data. How scores work