papers and frameworks

A curated repository linking state-of-the-art machine learning papers directly to their implementation frameworks.

Papers with Code bridges the gap between theoretical research and production-ready engineering by indexing 250,000+ machine learning papers with their corresponding GitHub repositories. It provides standardized benchmarks across 8,000+ tasks (such as image classification or natural language processing) to track the current state-of-the-art. By surfacing official codebases and community implementations in PyTorch, TensorFlow, and JAX, it eliminates the friction of reproducing complex neural architectures.

https://paperswithcode.com