# MatthieuCourbariaux/BinaryNet

Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1

Repository: https://github.com/MatthieuCourbariaux/BinaryNet
Canonical: https://ross.abutalabs.com/products/binarynet
Language: Python
License: BSD-3-Clause
License Family: permissive
Last push: 2018-11-28T21:45:38+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3860, "days_push": 2835, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1068, forks 339 (observed 2026-08-28T04:03:27.540290+00:00)

## What it is
BinaryNet is a research codebase for training deep neural networks whose weights and activations are constrained to +1 or -1, reproducing the experiments from the BinaryNet paper. It includes train-time benchmark code and run-time XNOR/baseline GPU kernels.

## Use cases
- train binary neural networks with +1/-1 weights and activations
- reproduce BinaryNet paper benchmark results
- run XNOR GPU kernels for fast binary network inference
- experiment with extremely low-bitwidth deep learning models
- study efficient neural network training techniques

## When to choose
- you need to reproduce or extend the BinaryNet paper's experiments
- you are researching binary/quantized neural networks
- you want reference XNOR GPU kernel implementations

## When to avoid
- you need a maintained production deep learning framework
- you want modern quantization support in PyTorch or TensorFlow
- you need up-to-date GPU compatibility and support

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, llm-training
- domain: deep-learning, machine-learning, artificial-intelligence
- platform: python
- tags: binary-neural-networks, bnn, xnor-kernels, research-code, model-compression, theano, gpu

## Member repositories
- MatthieuCourbariaux/BinaryNet (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:27.540290+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:54:37.064212+00:00, confidence not recorded.
  - readme: https://github.com/MatthieuCourbariaux/BinaryNet (fetched 2026-08-28T04:03:27.540290+00:00, sha 97dbf44861d3)
- Data as of 2026-08-30T08:39:29.467469+00:00.