Person_and_car_Detection_YOLOV5

System Setup for testing:

1) Install PyTorch as per your system requirement Link: https://pytorch.org/
(GPU and CUDA is recommended for High FPS on Video, else for images its fine.)
2) Create new Environment (pipenv or Conda)
And install requirements.txt file using below command
o pip install -r requirements.txt
3) Incase you want to test this on YouTube videos, then install below 2 libraries additionally.
o pip install youtube_dl
o pip install pafy
(System setup is ready and now your good to test)

Testing:

For images:
o python detect.py --source testimages/*.jpg --weights / best.pt
For YouTube Video:
o python detect.py --source https://youtu.be/QIjZn_fiS3M --weights best.pt --conf
0.25
“--source” can be anything from below:
◼ 0 # Webcam
◼ file.jpg # image
◼ file.mp4 # video
◼ path/ # directory
◼ path/*.jpg # glob
◼ 'https://youtu.be/NUsoVlDFqZg' # YouTube video
◼ 'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP stream

“--weights” by default, have I have kept best.pt