- Open up a terminal and change your directory to a directory that you want the projec to be in.
- Clone the project's GitHub here
git clone https://github.com/jwtanx/Rubbish_Collector .
- Clone the repo for yolov5
git clone https://github.com/ultralytics/yolov5
- Make sure you have python version 3.8 or above
If not you can install the python3.8 but take note
!!! DO NOT CHANGE THE PYTHON DEFAULT VERSION !!!
sudo apt-get update
sudo apt-get install python3.8 python3.8-dev python3.8-distutils python3.8-venv
- Create a virtual python virtual environement with >= Python 3.8
python3.8 -m venv jupiter
- Activate the virtual environment, when the environment is activated, you can see the (jupiter) on the left hand side of your command line
source jupiter/bin/activate
- Install the requirements, update your pip to the latest version if the installation stopped halfway
pip install -r requirements.txt
- !!! IMPORTANT : FIRST TIME SETUP ONLY !!!
python ./setup.py
If your donwload for the model failed, head to this Google Drive link and put the downloaded model pt file into this project directory.
- Start the inferencing on the robot camera
python ./yolov5/rubbish_collector.py --weights ./rubbish_detector.pt --source 2 --nosave
Try replacing the2
with0
or1
if the webcam is not working with2
When first time run, you might seerequirements: torch!=1.12.0,>=1.7.0 not found and is required by YOLOv5, attempting auto-update...
- To quit, make sure the Python Window is focused and hold the
q
key until the window is exited - It is recommended to deactivate the environment when you want are done with the application
Typedeactivate
to deactivate the envuronment before you quit the terminal
- Jupiter Robot we are working with
- Documentation for the robot ROS