Yolov5_ros
A ROS function package for PyTorch-YOLOv5 based on PyTorch -YOLOv5 is provided . This feature pack has been tested on Ubuntu 16.04 and Ubuntu 18.04.
Authors: Zhitao Zheng ([email protected])
Development environment:
- Ubuntu 16.04 / 18.04
- ROS kinetic/melodic
- Python>=3.6.0 environment, PyTorch>=1.7
Environment configuration installation steps:
Install Anaconda:
1. First download the corresponding installation package [Anaconda] (https://www.anaconda.com/products/individual#linux)
2. Then execute the script to install anaconda (the file name is the .sh file name corresponding to the download)
bash ~/Downloads/Anaconda3-2021.05-Linux-x86_64.sh
Install Pytorch:
1. First create a conda environment of python3.6 or above
conda create -n mypytorch python=3.8
2. Activate the created conda environment
conda activate mypytorch
3. Select the specified version on the PyTorch official website to install Pytorch
Install PyTorch: https://pytorch.org/get-started/locally/
conda install pytorch torchvision torchaudio cpuonly -c pytorch
Install Yolov5_ROS
cd /your/catkin_ws/src
git clone https://github.com/qq44642754a/Yolov5_ros.git
cd yolov5_ros/yolov5
sudo pip install -r requirements.txt
basic usage
- First, make sure to put your trained weights in the weights folder.
launch/yolo_v5.launchThe weight file `yolov5s.pt’ is used by default in the file. In addition, you need to additionally modify the name of your corresponding camera topic and whether to use the Cpu option in the launch file:
roslaunch yolov5_ros yolo_v5.launch
Special Instructions:
In view of the fact that many people have recently installed and configured a poorly configured yolov5 environment, I have configured a conda environment named yolov5 here.
Baidu network disk extraction code (pjgj).
After downloading, unzip the file and put it in the /anaconda3/envs/ folder, then run the following command to add it to the conda environment
conda config --add envs_dirs 下载好的yolov5文件夹的路径
finally passed the command
conda activate yolov5
To activate the conda environment
yolov5