yolov5目标检测模型的知识蒸馏（基于响应的蒸馏）

Last update: Dec 04, 2022

Overview

代码地址：

https://github.com/Sharpiless/yolov5-knowledge-distillation

教师模型：

python train.py --weights weights/yolov5m.pt \
        --cfg models/yolov5m.yaml --data data/voc.yaml --epochs 50 \
        --batch-size 8 --device 0 --hyp data/hyp.scratch.yaml

蒸馏训练：

python train.py --weights weights/yolov5s.pt \
        --cfg models/yolov5s.yaml --data data/voc.yaml --epochs 50 \
        --batch-size 8 --device 0 --hyp data/hyp.scratch.yaml \
        --t_weights yolov5m.pt --distill

训练参数:

--weights：预训练模型

--t_weights：教师模型权重

--distill：使用知识蒸馏进行训练

--dist_loss：l2或者kl

--temperature：使用知识蒸馏时的温度

使用《Object detection at 200 Frames Per Second》中的损失

这篇文章分别对这几个损失函数做出改进，具体思路为只有当teacher network的objectness value高时，才学习bounding box坐标和class probabilities。

实验结果：

这里假设VOC2012中新增加的数据为无标签数据（2k张）。

教师模型	训练方法	蒸馏损失	P	R	mAP50
无	正常训练	不使用	0.7756	0.7115	0.7609
Yolov5l	output based	l2	0.7585	0.7198	0.7644
Yolov5l	output based	KL	0.7417	0.7207	0.7536
Yolov5m	output based	l2	0.7682	0.7436	0.7976
Yolov5m	output based	KL	0.7731	0.7313	0.7931

参数和细节正在完善，支持KL散度、L2 logits损失和Sigmoid蒸馏损失等

1. 正常训练：

2. L2蒸馏损失：

我的公众号：

关于作者

B站：https://space.bilibili.com/470550823

CSDN：https://blog.csdn.net/weixin_44936889

AI Studio：https://aistudio.baidu.com/aistudio/personalcenter/thirdview/67156

Github：https://github.com/Sharpiless

yolov5目标检测模型的知识蒸馏（基于响应的蒸馏）

Related tags

Overview

代码地址：

教师模型：

蒸馏训练：

训练参数:

实验结果：

1. 正常训练：

2. L2蒸馏损失：

我的公众号：

关于作者

Owner

Homepage of paper: Paint Transformer: Feed Forward Neural Painting with Stroke Prediction, ICCV 2021.

Pytorch-Swin-Unet-V2 - a modified version of Swin Unet based on Swin Transfomer V2

Cascaded Deep Video Deblurring Using Temporal Sharpness Prior and Non-local Spatial-Temporal Similarity

Code repository for the paper Computer Vision User Entity Behavior Analytics

A collection of awesome resources image-to-image translation.

Model Serving Made Easy

Point detection through multi-instance deep heatmap regression for sutures in endoscopy

Official PyTorch implementation and pretrained models of the paper Self-Supervised Classification Network

HHP-Net: A light Heteroscedastic neural network for Head Pose estimation with uncertainty

Pairwise Learning for Neural Link Prediction for OGB (PLNLP-OGB)

[NeurIPS 2021] A weak-shot object detection approach by transferring semantic similarity and mask prior.

Easily benchmark PyTorch model FLOPs, latency, throughput, max allocated memory and energy consumption

PyTorch implementation of Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy

This repository is maintained for the scientific paper tittled " Study of keyword extraction techniques for Electric Double Layer Capacitor domain using text similarity indexes: An experimental analysis "

Lightwood is Legos for Machine Learning.

PyTorch code for our paper "Attention in Attention Network for Image Super-Resolution"

Supervised multi-SNE (S-multi-SNE): Multi-view visualisation and classification

PyTorch Implementation of DiffGAN-TTS: High-Fidelity and Efficient Text-to-Speech with Denoising Diffusion GANs

Minimal But Practical Image Classifier Pipline Using Pytorch, Finetune on ResNet18, Got 99% Accuracy on Own Small Datasets.

An original implementation of "MetaICL Learning to Learn In Context" by Sewon Min, Mike Lewis, Luke Zettlemoyer and Hannaneh Hajishirzi