donglixp / apex

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool


This repository holds NVIDIA-maintained utilities to streamline mixed precision and distributed training in Pytorch. Some of the code here will be included in upstream Pytorch eventually. The intention of Apex is to make up-to-date utilities available to users as quickly as possible.

Full API Documentation:


1. Mixed Precision

amp: Automatic Mixed Precision

apex.amp is a tool designed for ease of use and maximum safety in FP16 training. All potentially unsafe ops are performed in FP32 under the hood, while safe ops are performed using faster, Tensor Core-friendly FP16 math. amp also automatically implements dynamic loss scaling.

The intention of amp is to be the "on-ramp" to easy FP16 training: achieve all the numerical stability of full FP32 training, with most of the performance benefits of full FP16 training.

Python Source and API Documentation


apex.FP16_Optimizer wraps an existing Python optimizer and automatically implements master parameters and static or dynamic loss scaling under the hood.

The intention of FP16_Optimizer is to be the "highway" for FP16 training: achieve most of the numerically stability of full FP32 training, and almost all the performance benefits of full FP16 training.

API Documentation

Python Source

Simple examples with FP16_Optimizer

Imagenet with FP16_Optimizer

word_language_model with FP16_Optimizer

The Imagenet and word_language_model directories also contain examples that show manual management of master parameters and static loss scaling.

These manual examples illustrate what sort of operations amp and FP16_Optimizer are performing automatically.

2. Distributed Training

apex.parallel.DistributedDataParallel is a module wrapper, similar to torch.nn.parallel.DistributedDataParallel. It enables convenient multiprocess distributed training, optimized for NVIDIA's NCCL communication library.

API Documentation

Python Source


The Imagenet with FP16_Optimizer mixed precision examples also demonstrate apex.parallel.DistributedDataParallel.

Synchronized Batch Normalization

apex.parallel.SyncBatchNorm extends torch.nn.modules.batchnorm._BatchNorm to support synchronized BN. It reduces stats across processes during multiprocess distributed data parallel training. Synchronous Batch Normalization has been used in cases where only very small number of mini-batch could be fit on each GPU. All-reduced stats boost the effective batch size for sync BN layer to be the total number of mini-batches across all processes. It has improved the converged accuracy in some of our research models.


Python 3


PyTorch 0.4 or newer. We recommend to use the latest stable release, obtainable from We also test against the latest master branch, obtainable from
If you have any problems building, please file an issue.

Quick Start


To build the extension run

python install

in the root directory of the cloned repository.

To use the extension

import apex

CUDA/C++ extension

Apex contains optional CUDA/C++ extensions, installable via

python install [--cuda_ext] [--cpp_ext]

Currently, --cuda_ext enables

  • Fused kernels that improve the performance and numerical stability of apex.parallel.SyncBatchNorm.
  • Fused kernels required to use apex.optimizers.FusedAdam.
  • Fused kernels required to use 'apex.normalization.FusedLayerNorm'.

--cpp_ext enables

  • C++-side flattening and unflattening utilities that reduce the CPU overhead of apex.parallel.DistributedDataParallel.

Windows support

Windows support is experimental, and Linux is recommended. However, since Apex could be Python-only, there's a good chance the Python-only features "just works" the same way as Linux. If you installed Pytorch in a Conda environment, make sure to install Apex in that same environment.

ezoic increase your site revenue


A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

License:BSD 3-Clause "New" or "Revised" License


Language:Python 79.0%Language:Cuda 17.2%Language:C++ 3.7%Language:Shell 0.1%