Paddle pit - Rethinking Spatial Dimensions of Vision Transformers

Overview

基于Paddle实现PiT ——Rethinking Spatial Dimensions of Vision Transformers,arxiv

  • 官方原版代码(基于PyTorch)pit.

  • 本项目基于 PaddleViT 实现,在其基础上与原版代码实现了更进一步的对齐,并通过完整训练与测试完成对pit_ti模型的复现.

1. 简介

从CNN的成功设计原理出发,作者研究了空间尺寸转换的作用及其在基于Transformer的体系结构上的有效性。

具体来说,类似于CNN的降维原则(随着深度的增加,传统的CNN会增加通道尺寸并减小空间尺寸),作者用实验表明了这同样有利于Transformer的性能提升,并提出了基于池化的Vision Transformer,即PiT(模型示意图如下)。

drawing

PiT 模型示意图

2. 数据集和复现精度

数据集

原文使用的为ImageNet-1k 2012(ILSVRC2012),共1000类,训练集/测试集图片分布:1281167/50000,数据集大小为144GB。

本项目使用的为官方推荐的图片压缩过的更轻量的Light_ILSVRC2012,数据集大小为65GB。其在AI Studio上的地址为:Light_ILSVRC2012_part_0.tarLight_ILSVRC2012_part_1.tar

复现精度

Model 目标精度[email protected] 实现精度[email protected] Image Size batch_size Crop_pct epoch #Params
pit_ti 73.0 73.01 224 256*4GPUs 0.9 300
(+10 COOLDOWN)
4.8M

【注】上表中的实现精度在原版ILSVRC2012验证集上测试得到。 值得一提的是,本项目在Light_ILSVRC2012的验证集上的Validation [email protected]达到了73.17

本项目训练得到的最佳模型参数与训练日志log均存放于output文件夹下。

日志文件说明

本项目通过AI Studio的脚本任务运行,中途中断了4次,因此共有5个日志文件。为了方便检阅,本人手动将log命名为log_开始epoch-结束epoch.txt格式。具体来说:

  • output/log_1-76.txt:epoch1~epoch76。这一版代码定义每10个epoch保存一次模型权重,每2个epoch验证一次,同时若验证精度高于历史精度,则保存为Best_PiT.pdparams,因此在epoch76训练结束但还未验证的时候中断,下一次的训练只能从验证精度最高的epoch74继续训练。

  • output/log_75-142.txt:epoch75~epoch142。从这一版代码开始,新增了每次训练之后都保存一下模型参数为PiT-Latest.pdparams,这样无论哪个epoch训练中断都可以继续训练啦。

  • output/log_143-225.txt:epoch143~epoch225。

  • output/log_226-303.txt:epoch226~epoch303。

  • output/log_304-310.txt:epoch304~epoch310。

  • output/log_eval.txt:使用训练得到的最好模型(epoch308)在原版ILSVRC2012验证集上验证日志。

3. 准备环境

推荐环境配置:

本人环境配置:

  • 硬件:Tesla V100 * 4(由衷感谢百度飞桨平台提供高性能算力支持)

  • PaddlePaddle==2.2.1

  • Python==3.7

4. 快速开始

本项目现已通过脚本任务形式部署到AI Studio上,您可以选择fork下来直接运行sh run.sh,数据集处理等脚本均已部署好。链接:paddle_pit

或者您也可以git本repo在本地运行:

第一步:克隆本项目

git clone https://github.com/hatimwen/paddle_pit.git
cd paddle_pit

第二步:修改参数

请根据实际情况,修改scripts路径下的脚本内容(如:gpu,数据集路径data_path,batch_size等)。

第三步:验证模型

多卡请运行:

sh scripts/run_eval_multi.sh

单卡请运行:

sh scripts/run_eval.sh

第四步:训练模型

多卡请运行:

sh scripts/run_train_multi.sh

单卡请运行:

sh scripts/run_train.sh

第五步:验证预测

python predict.py \
-pretrained='output/Best_PiT' \
-img_path='images/ILSVRC2012_val_00004506.JPEG'

验证图片(类别:藏獒, id: 244)

输出结果为:

class_id: 244, prob: 9.12291145324707

对照ImageNet类别id(ImageNet数据集编号对应的类别内容),可知244为藏獒,预测结果正确。

5.代码结构

|-- paddle_pit
    |-- output              # 日志及模型文件
    |-- configs             # 参数
        |-- pit_ti.yaml
    |-- datasets
        |-- ImageNet1K      # 数据集路径
    |-- scripts             # 运行脚本
        |-- run_train.sh
        |-- run_train_multi.sh
        |-- run_eval.sh
        |-- run_eval_multi.sh
    |-- augment.py          # 数据增强
    |-- config.py           # 最底层配置文件
    |-- datasets.py         # dataset与dataloader
    |-- droppath.py         # droppath定义
    |-- losses.py           # loss定义
    |-- main_multi_gpu.py   # 多卡训练测试代码
    |-- main_single_gpu.py  # 单卡训练测试代码
    |-- mixup.py            # mixup定义
    |-- model_ema.py        # EMA定义
    |-- pit.py              # pit模型结构定义
    |-- random_erasing.py   # random_erasing定义
    |-- regnet.py           # 教师模型定义(本项目并未对此验证,仅作保留)
    |-- transforms.py       # RandomHorizontalFlip定义
    |-- utils.py            # CosineLRScheduler及AverageMeter定义
    |-- README.md
    |-- requirements.txt

6. 参考及引用

@InProceedings{Yuan_2021_ICCV,
    author    = {Yuan, Li and Chen, Yunpeng and Wang, Tao and Yu, Weihao and Shi, Yujun and Jiang, Zi-Hang and Tay, Francis E.H. and Feng, Jiashi and Yan, Shuicheng},
    title     = {Tokens-to-Token ViT: Training Vision Transformers From Scratch on ImageNet},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    month     = {October},
    year      = {2021},
    pages     = {558-567}
}

最后,非常感谢百度举办的飞桨论文复现挑战赛(第五期)让本人对Paddle理解更加深刻。 同时也非常感谢朱欤老师团队用Paddle实现的PaddleViT,本项目大部分代码都是从中copy来的,而仅仅实现了其与原版代码训练步骤的进一步对齐与完整的训练过程,但本人也同样受益匪浅! ♥️

Contact

Owner
Hongtao Wen
Hongtao Wen
Towards Fine-Grained Reasoning for Fake News Detection

FinerFact This is the PyTorch implementation for the FinerFact model in the AAAI 2022 paper Towards Fine-Grained Reasoning for Fake News Detection (Ar

Ahren_Jin 15 Dec 15, 2022
PaddleRobotics is an open-source algorithm library for robots based on Paddle, including open-source parts such as human-robot interaction, complex motion control, environment perception, SLAM positioning, and navigation.

简体中文 | English PaddleRobotics paddleRobotics是基于paddle的机器人开源算法库集,包括人机交互、复杂运动控制、环境感知、slam定位导航等开源算法部分。 人机交互 主动多模交互技术TFVT-HRI 主动多模交互技术是通过视觉、语音、触摸传感器等输入机器人

185 Dec 26, 2022
kullanışlı ve işinizi kolaylaştıracak bir araç

Hey merhaba! işte çok sorulan sorularının cevabı ve sorunlarının çözümü; Soru= İçinde var denilen birçok şeyi göremiyorum bunun sebebi nedir? Cevap= B

Sexettin 16 Dec 17, 2022
A GPU-optional modular synthesizer in pytorch, 16200x faster than realtime, for audio ML researchers.

torchsynth The fastest synth in the universe. Introduction torchsynth is based upon traditional modular synthesis written in pytorch. It is GPU-option

torchsynth 229 Jan 02, 2023
Learning Saliency Propagation for Semi-supervised Instance Segmentation

Learning Saliency Propagation for Semi-supervised Instance Segmentation PyTorch Implementation This repository contains: the PyTorch implementation of

Berkeley DeepDrive 68 Oct 18, 2022
Training and Evaluation Code for Neural Volumes

Neural Volumes This repository contains training and evaluation code for the paper Neural Volumes. The method learns a 3D volumetric representation of

Meta Research 370 Dec 08, 2022
Tensorflow Repo for "DeepGCNs: Can GCNs Go as Deep as CNNs?"

DeepGCNs: Can GCNs Go as Deep as CNNs? In this work, we present new ways to successfully train very deep GCNs. We borrow concepts from CNNs, mainly re

Guohao Li 612 Nov 15, 2022
Machine learning, in numpy

numpy-ml Ever wish you had an inefficient but somewhat legible collection of machine learning algorithms implemented exclusively in NumPy? No? Install

David Bourgin 11.6k Dec 30, 2022
GeoMol: Torsional Geometric Generation of Molecular 3D Conformer Ensembles

GeoMol: Torsional Geometric Generation of Molecular 3D Conformer Ensembles This repository contains a method to generate 3D conformer ensembles direct

127 Dec 20, 2022
Official PyTorch implementation of RobustNet (CVPR 2021 Oral)

RobustNet (CVPR 2021 Oral): Official Project Webpage Codes and pretrained models will be released soon. This repository provides the official PyTorch

Sungha Choi 173 Dec 21, 2022
Python3 / PyTorch implementation of the following paper: Fine-grained Semantics-aware Representation Enhancement for Self-supervisedMonocular Depth Estimation. ICCV 2021 (oral)

FSRE-Depth This is a Python3 / PyTorch implementation of FSRE-Depth, as described in the following paper: Fine-grained Semantics-aware Representation

77 Dec 28, 2022
A library that can print Python objects in human readable format

objprint A library that can print Python objects in human readable format Install pip install objprint Usage op Use op() (or objprint()) to print obj

319 Dec 25, 2022
We propose a new method for effective shadow removal by regarding it as an exposure fusion problem.

Auto-exposure fusion for single-image shadow removal We propose a new method for effective shadow removal by regarding it as an exposure fusion proble

Qing Guo 146 Dec 31, 2022
Human pose estimation from video plays a critical role in various applications such as quantifying physical exercises, sign language recognition, and full-body gesture control.

Pose Detection Project Description: Human pose estimation from video plays a critical role in various applications such as quantifying physical exerci

Hassan Shahzad 2 Jan 17, 2022
A Python package for time series augmentation

tsaug tsaug is a Python package for time series augmentation. It offers a set of augmentation methods for time series, as well as a simple API to conn

Arundo Analytics 278 Jan 01, 2023
Structured Data Gradient Pruning (SDGP)

Structured Data Gradient Pruning (SDGP) Weight pruning is a technique to make Deep Neural Network (DNN) inference more computationally efficient by re

Bradley McDanel 10 Nov 11, 2022
Repositorio oficial del curso IIC2233 Programación Avanzada 🚀✨

IIC2233 - Programación Avanzada Evaluación Las evaluaciones serán efectuadas por medio de actividades prácticas en clases y tareas. Se calculará la no

IIC2233 @ UC 0 Dec 15, 2022
Pytorch implementation of our paper under review — Lottery Jackpots Exist in Pre-trained Models

Lottery Jackpots Exist in Pre-trained Models (Paper Link) Requirements Python = 3.7.4 Pytorch = 1.6.1 Torchvision = 0.4.1 Reproduce the Experiment

Yuxin Zhang 27 Jun 28, 2022
Learning Super-Features for Image Retrieval

Learning Super-Features for Image Retrieval This repository contains the code for running our FIRe model presented in our ICLR'22 paper: @inproceeding

NAVER 101 Dec 28, 2022
ADSPM: Attribute-Driven Spontaneous Motion in Unpaired Image Translation

ADSPM: Attribute-Driven Spontaneous Motion in Unpaired Image Translation This repository provides a PyTorch implementation of ADSPM. Requirements Pyth

24 Jul 24, 2022