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

Official PyTorch Implementation for InfoSwap: Information Bottleneck Disentanglement for Identity Swapping

Pytorch implementation of the AAAI 2022 paper "Cross-Domain Empirical Risk Minimization for Unbiased Long-tailed Classification"

Learning with Noisy Labels via Sparse Regularization, ICCV2021

Doubly Robust Off-Policy Evaluation for Ranking Policies under the Cascade Behavior Model

This repository contains answers of the Shopify Summer 2022 Data Science Intern Challenge.

Automatic packaging of the open-composite libs for OvGME

Unrestricted Facial Geometry Reconstruction Using Image-to-Image Translation

OHLC Average Prediction of Apple Inc. Using LSTM Recurrent Neural Network

Use VITS and Opencpop to develop singing voice synthesis; Maybe it will VISinger.

Add gui for YoloV5 using PyQt5

Edge-oriented Convolution Block for Real-time Super Resolution on Mobile Devices, ACM Multimedia 2021

The source code of "SIDE: Center-based Stereo 3D Detector with Structure-aware Instance Depth Estimation", accepted to WACV 2022.

Source code for Fathony, Sahu, Willmott, & Kolter, "Multiplicative Filter Networks", ICLR 2021.

realsense d400 -> jpg + csv

Implementation of ECCV20 paper: the devil is in classification: a simple framework for long-tail object detection and instance segmentation

Zen-NAS: A Zero-Shot NAS for High-Performance Deep Image Recognition

A scikit-learn compatible neural network library that wraps PyTorch

A highly efficient, fast, powerful and light-weight anime downloader and streamer for your favorite anime.

Flask101 - FullStack Web Development with Python & JS - From TAQWA

List of papers, code and experiments using deep learning for time series forecasting