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

Python package for multiple object tracking research with focus on laboratory animals tracking.

Official Datasets and Implementation from our Paper "Video Class Agnostic Segmentation in Autonomous Driving".

The lightweight PyTorch wrapper for high-performance AI research. Scale your models, not the boilerplate.

Depth image based mouse cursor visual haptic

DeepProbLog is an extension of ProbLog that integrates Probabilistic Logic Programming with deep learning by introducing the neural predicate.

This is the official implementation of Elaborative Rehearsal for Zero-shot Action Recognition (ICCV2021)

This is an easy python software which allows to sort images with faces by gender and after by age.

A simple Python configuration file operator.

A minimalist tool to display a network graph.

The (Official) PyTorch Implementation of the paper "Deep Extraction of Manga Structural Lines"

Tweesent-back - Tweesent backend uses fastAPI as the web framework

Episodic Transformer (E.T.) is a novel attention-based architecture for vision-and-language navigation. E.T. is based on a multimodal transformer that encodes language inputs and the full episode history of visual observations and actions.

Data, model training, and evaluation code for "PubTables-1M: Towards a universal dataset and metrics for training and evaluating table extraction models".

A set of tests for evaluating large-scale algorithms for Wasserstein-2 transport maps computation.

DiSECt: Differentiable Simulator for Robotic Cutting

Kaggle competition: Springleaf Marketing Response

A object detecting neural network powered by the yolo architecture and leveraging the PyTorch framework and associated libraries.

Code of the paper "Shaping Visual Representations with Attributes for Few-Shot Learning (ASL)".

Implementation of the HMAX model of vision in PyTorch

WSDM‘2022: Knowledge Enhanced Sports Game Summarization