资源分类
》 资源 》算法》ChainerMN

ChainerMN

2019-09-17 | |  70 |   0 |   0

Build Status Documentation Status PyPI License: MIT

ChainerMN: Distributed Deep Learning with Chainer

Documentation | Installation | Examples | Release Notes

ChainerMN is an additional package for Chainer, a flexible deep learning framework. ChainerMN enables multi-node distributed deep learning with the following features:

  • Scalable --- it makes full use of the latest technologies such as NVIDIA NCCL and CUDA-Aware MPI,

  • Flexible --- even dynamic neural networks can be trained in parallel thanks to Chainer's flexibility, and

  • Easy --- minimal changes to existing user code are required.

This blog post provides our benchmark results using up to 128 GPUs.

上一篇：VIN_PyTorch_Visdom

下一篇：LearningToCompare-Pytorch

用户评价
全部评价

热门资源

  • Keras-ResNeXt

    Keras ResNeXt Implementation of ResNeXt models...

  • seetafaceJNI

    项目介绍 基于中科院seetaface2进行封装的JAVA...

  • spark-corenlp

    This package wraps Stanford CoreNLP annotators ...

  • capsnet-with-caps...

    CapsNet with capsule-wise convolution Project ...

  • inferno-boilerplate

    This is a very basic boilerplate example for pe...