Source of original project: DAMO-YOLO
PapersWithCode: DAMO-YOLO: A Report on Real-Time Object Detection Design
NOTE: If you don't have Docker, install it using documentation!
NOTE: Available only for Windows and Linux!
NOTE: If the inference of the model causes an error for you, it may be worth updating the paths in
download_model.py
!
git clone https://github.com/AlibekovMurad5202/DLOps-practice.git && cd DLOps-practice
docker build -t damoyolo .
Ubuntu:
docker run -it -v "$(pwd)":/DLOps damoyolo
Windows:
docker run -it -v <path_to_DLOps-practice>:/DLOps damoyolo
Example (for Windows):
docker run -it -v "C:\Users\murad\Desktop\ITMM\tmp\DLOps-practice":/DLOps damoyolo
conda activate yolo_env && source DAMO-YOLO-env/bin/activate
cd DLOps && python3 download_model.py -m "damoyolo_tinynasL25_S.pth"
sed -i -e 's/\r$//' run.sh
chmod +x run.sh
./run.sh
exit
docker ps -a
docker stop CONTAINER_ID
docker rm CONTAINER_ID