MXNet logo

MXNet

Allows mixing symbolic and imperative programming to maximize efficiency and productivity with automatic parallelization of operations.

Made by The Apache Software Foundation

  • development

  • deep-learning

  • Machine Learning

  • Python

What is MXNet?

MXNet is a deep learning framework that offers a unique blend of efficiency and flexibility. Its core features a dynamic dependency scheduler that automatically parallelizes both symbolic and imperative operations, optimizing performance on the fly. MXNet's graph optimization layer further enhances symbolic execution, delivering fast and memory-efficient computations. Designed to be portable and lightweight, the framework scales effectively across multiple GPUs and machines, empowering users to tackle a wide range of deep learning challenges

Highlights

  • Hybrid programming model: Allows mixing of symbolic and imperative programming styles to maximize efficiency and productivity
  • Dynamic dependency scheduling: Automatically parallelizes both symbolic and imperative operations for optimized performance
  • Graph optimization: Enhances symbolic execution for fast and memory-efficient computations
  • Portability and scalability: Runs efficiently on multiple GPUs and machines

Platforms

  • Mac
  • Linux
  • Web
  • Windows

Languages

  • English

Social