Simple tutorials on Pytorch DDP training

Overview

pytorch-distributed-training

Distribute Dataparallel (DDP) Training on Pytorch

Features

Good Notes

分享一些网上优质的笔记

TODO

  • 完成DP和DDP源码解读笔记(当前进度50%)
  • 修改代码细节, 复现实验结果

Quick start

想直接运行查看结果的可以执行以下命令, 注意一定要用--ip--port来指定主机的ip地址以及空闲的端口,否则可能无法运行

$ python dataparallel.py --gpu 0,1,2,3
$ CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch --nproc_per_node=4 distributed.py
$ CUDA_VISIBLE_DEVICES=0,1,2,3 python distributed_mp.py
$ CUDA_VISIBLE_DEVICES=0,1,2,3 python distributed_apex.py
  • --ip=str, e.g --ip='10.24.82.10' 来指定主进程的ip地址

  • --port=int, e.g --port=23456 来指定启动端口号

  • --batch_size=int, e.g --batch_size=128 设定训练batch_size

  • distributed_gradient_accumulation.py

$ CUDA_VISIBLE_DEVICES=0,1,2,3 python distributed_apex.py
  • --ip=str, e.g --ip='10.24.82.10' 来指定主进程的ip地址
  • --port=int, e.g --port=23456 来指定启动端口号
  • --grad_accu_steps=int, e.g --grad_accu_steps=4' 来指定gradient_step

Comparison

结果不够准确,GPU状态不同结果可能差异较大

默认情况下都使用SyncBatchNorm, 这会导致执行速度变慢一些,因为需要增加进程之间的通讯来计算BatchNorm, 但有利于保证准确率

Concepts

  • apex
  • DP: DataParallel
  • DDP: DistributedDataParallel

Environments

  • 4 × 2080Ti
model dataset training method time(seconds/epoch) Top-1 accuracy
resnet18 cifar100 DP 20s
resnet18 cifar100 DP+apex 18s
resnet18 cifar100 DDP 16s
resnet18 cifar100 DDP+apex 14.5s

Basic Concept

  • group: 表示进程组,默认情况下只有一个进程组。
  • world size: 全局进程个数
    • 比如16张卡单卡单进程: world size = 16
    • 8卡单进程: world size = 1
    • 只有当连接的进程数等于world size, 程序才会执行
  • rank: 进程序号,用于进程间通讯,表示进程优先级,rank=0表示主进程
  • local_rank: 进程内,GPU编号,非显示参数,由torch.distributed.launch内部指定,rank=3, local_rank=0 表示第3个进程的第1GPU

Usage 单机多卡

1. 获取当前进程的index

pytorch可以通过torch.distributed.lauch启动器,在命令行分布式地执行.py文件, 在执行的过程中会将当前进程的index通过参数传递给python

import argparse
parser = argparse.ArgumentParser()
parser.add_argument('--local_rank', default=-1, type=int,
                    help='node rank for distributed training')
args = parser.parse_args()
print(args.local_rank)

2. 定义 main_worker 函数

主要的训练流程都写在main_worker函数中,main_worker需要接受三个参数(最后一个参数optional):

def main_worker(local_rank, nprocs, args):
    training...
  • local_rank: 接受当前进程的rank值,在一机多卡的情况下对应使用的GPU号
  • nprocs: 进程数量
  • args: 自己定义的额外参数

main_worker,相当于你每个进程需要运行的函数(每个进程执行的函数内容是一致的,只不过传入的local_rank不一样)

3. main_worker函数中的整体流程

main_worker函数中完整的训练流程

import torch
import torch.distributed as dist
import torch.backends.cudnn as cudnn
def main_worker(local_rank, nprocs, args):
    args.local_rank = local_rank
    # 分布式初始化,对于每个进程来说,都需要进行初始化
    cudnn.benchmark = True
    dist.init_process_group(backend='nccl', init_method='tcp://ip:port', world_size=nprocs, rank=local_rank)
    # 模型、损失函数、优化器定义
    model = ...
    criterion = ...
    optimizer = ...
    # 设置进程对应使用的GPU
    torch.cuda.set_device(local_rank)
    model.cuda(local_rank)
    # 使用分布式函数定义模型
    model = model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[local_rank])
    
    # 数据集的定义,使用 DistributedSampler
    mini_batch_size = batch_size / nprocs # 手动划分 batch_size to mini-batch_size
    train_dataset = ...
    train_sampler = torch.utils.data.distributed.DistributedSampler(train_dataset)
    trainloader = torch.utils.data.DataLoader(train_dataset, batch_size=mini_batch_size, num_workers=..., pin_memory=..., 
                                              sampler=train_sampler)
    
    test_dataset = ...
    test_sampler = torch.utils.data.distributed.DistributedSampler(test_dataset)
    testloader = torch.utils.data.DataLoader(train_dataset, batch_size=mini_batch_size, num_workers=..., pin_memory=..., 
                                             sampler=test_sampler) 
    
    # 正常的 train 流程
    for epoch in range(300):
       model.train()
       for batch_idx, (images, target) in enumerate(trainloader):
          images = images.cuda(non_blocking=True)
          target = target.cuda(non_blocking=True)
          ...
          pred = model(images)
          loss = loss_function(pred, target)
          ...
          optimizer.zero_grad()
          loss.backward()
          optimizer.step()

4. 定义main函数

import argparse
import torch
parser = argparse.ArgumentParser(description='PyTorch ImageNet Training')
parser.add_argument('--local_rank', default=-1, type=int, help='node rank for distributed training')
parser.add_argument('--batch_size','--batch-size', default=256, type=int)
parser.add_argument('--lr', default=0.1, type=float)

def main_worker(local_rank, nprocs, args):
    ...

def main():
    args = parser.parse_args()
    args.nprocs = torch.cuda.device_count()
    # 执行 main_worker
    main_worker(args.local_rank, args.nprocs, args)

if __name__ == '__main__':
    main()

5. Command Line 启动

$ CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch --nproc_per_node=4 distributed.py
  • --ip=str, e.g --ip='10.24.82.10' 来指定主进程的ip地址
  • --port=int, e.g --port=23456 来指定启动端口号

参数说明:

  • --nnodes 表示机器的数量
  • --node_rank 表示当前的机器
  • --nproc_per_node 表示每台机器上的进程数量

参考 distributed.py

6. torch.multiprocessing

使用torch.multiprocessing来解决进程自发控制可能产生问题,这种方式比较稳定,推荐使用

import argparse
import torch
import torch.multiprocessing as mp

parser = argparse.ArgumentParser(description='PyTorch ImageNet Training')
parser.add_argument('--local_rank', default=-1, type=int, help='node rank for distributed training')
parser.add_argument('--batch_size','--batch-size', default=256, type=int)
parser.add_argument('--lr', default=0.1, type=float)

def main_worker(local_rank, nprocs, args):
    ...

def main():
    args = parser.parse_args()
    args.nprocs = torch.cuda.device_count()
    # 将 main_worker 放入 mp.spawn 中
    mp.spawn(main_worker, nprocs=args.nprocs, args=(args.nprocs, args))

if __name__ == '__main__':
    main()

参考 distributed_mp.py 启动方式如下:

$ CUDA_VISIBLE_DEVICES=0,1,2,3 python distributed_mp.py
  • --ip=str, e.g --ip='10.24.82.10' 来指定主进程的ip地址
  • --port=int, e.g --port=23456 来指定启动端口号

Implemented Work

参考的文章如下(如果有文章没有引用,但是内容差不多的,可以提issue给我,我会补上,实在抱歉):

Owner
Ren Tianhe
Ren Tianhe
Scalable Multi-Agent Reinforcement Learning

Scalable Multi-Agent Reinforcement Learning 1. Featured algorithms: Value Function Factorization with Variable Agent Sub-Teams (VAST) [1] 2. Implement

3 Aug 02, 2022
Official Repo for Ground-aware Monocular 3D Object Detection for Autonomous Driving

Visual 3D Detection Package: This repo aims to provide flexible and reproducible visual 3D detection on KITTI dataset. We expect scripts starting from

Yuxuan Liu 305 Dec 19, 2022
MXNet implementation for: Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution

Octave Convolution MXNet implementation for: Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution Imag

Meta Research 549 Dec 28, 2022
An automated algorithm to extract the linear blend skinning (LBS) from a set of example poses

Dem Bones This repository contains an implementation of Smooth Skinning Decomposition with Rigid Bones, an automated algorithm to extract the Linear B

Electronic Arts 684 Dec 26, 2022
Chinese Advertisement Board Identification(Pytorch)

Chinese-Advertisement-Board-Identification. We use YoloV5 to extract the ROI of the location of the chinese word. Next, we sort the bounding box and recognize every chinese words which we extracted.

Li-Wei Hsiao 12 Jul 21, 2022
YouRefIt: Embodied Reference Understanding with Language and Gesture

YouRefIt: Embodied Reference Understanding with Language and Gesture YouRefIt: Embodied Reference Understanding with Language and Gesture by Yixin Che

16 Jul 11, 2022
How the Deep Q-learning method works and discuss the new ideas that makes the algorithm work

Deep Q-Learning Recommend papers The first step is to read and understand the method that you will implement. It was first introduced in a 2013 paper

1 Jan 25, 2022
This is the solution for 2nd rank in Kaggle competition: Feedback Prize - Evaluating Student Writing.

Feedback Prize - Evaluating Student Writing This is the solution for 2nd rank in Kaggle competition: Feedback Prize - Evaluating Student Writing. The

Udbhav Bamba 41 Dec 14, 2022
Python Assignments for the Deep Learning lectures by Andrew NG on coursera with complete submission for grading capability.

Python Assignments for the Deep Learning lectures by Andrew NG on coursera with complete submission for grading capability.

Utkarsh Agiwal 1 Feb 03, 2022
Shape-aware Semi-supervised 3D Semantic Segmentation for Medical Images

SASSnet Code for paper: Shape-aware Semi-supervised 3D Semantic Segmentation for Medical Images(MICCAI 2020) Our code is origin from UA-MT You can fin

klein 125 Jan 03, 2023
Regularized Frank-Wolfe for Dense CRFs: Generalizing Mean Field and Beyond

CRF - Conditional Random Fields A library for dense conditional random fields (CRFs). This is the official accompanying code for the paper Regularized

Đ.Khuê Lê-Huu 21 Nov 26, 2022
deep learning model with only python and numpy with test accuracy 99 % on mnist dataset and different optimization choices

deep_nn_model_with_only_python_100%_test_accuracy deep learning model with only python and numpy with test accuracy 99 % on mnist dataset and differen

0 Aug 28, 2022
A curated list of awesome papers for Semantic Retrieval (TOIS Accepted: Semantic Models for the First-stage Retrieval: A Comprehensive Review).

A curated list of awesome papers for Semantic Retrieval (TOIS Accepted: Semantic Models for the First-stage Retrieval: A Comprehensive Review).

Yinqiong Cai 189 Dec 28, 2022
Small-bets - Ergodic Experiment With Python

Ergodic Experiment Based on this video. Run this experiment with this command: p

Michael Brant 3 Jan 11, 2022
A Japanese Medical Information Extraction Toolkit

JaMIE: a Japanese Medical Information Extraction toolkit Joint Japanese Medical Problem, Modality and Relation Recognition The Train/Test phrases requ

7 Dec 12, 2022
Object Depth via Motion and Detection Dataset

ODMD Dataset ODMD is the first dataset for learning Object Depth via Motion and Detection. ODMD training data are configurable and extensible, with ea

Brent Griffin 172 Dec 21, 2022
Official respository for "Modeling Defocus-Disparity in Dual-Pixel Sensors", ICCP 2020

Official respository for "Modeling Defocus-Disparity in Dual-Pixel Sensors", ICCP 2020 BibTeX @INPROCEEDINGS{punnappurath2020modeling, author={Abhi

Abhijith Punnappurath 22 Oct 01, 2022
for a paper about leveraging discourse markers for training new models

TSLM-DISCOURSE-MARKERS Scope This repository contains: (1) Code to extract discourse markers from wikipedia (TSA). (1) Code to extract significant dis

International Business Machines 6 Nov 02, 2022
It helps user to learn Pick-up lines and share if he has a better one

Pick-up-Lines-Generator(Open Source) It helps user to learn Pick-up lines Share and Add one or many to the DataBase Unique SQLite DataBase AI Undercon

knock_nott 0 May 04, 2022
GeoTransformer - Geometric Transformer for Fast and Robust Point Cloud Registration

Geometric Transformer for Fast and Robust Point Cloud Registration PyTorch imple

Zheng Qin 220 Jan 05, 2023