Hello World! This reposotory provied code for training and inference of player tracking!

0. Before running install Docker

To run a raw video file for tracking, run the following:

1. git clone https://github.com/oljikeboost/Tracking
2. cd Tracking
3. nvidia-docker build -t tracking . (here we build the docker image from which we create container)
4. nvidia-docker run -v path/to/directory/with/video/and/ocr/GAME_NAME:/home/user/data/GAME_NAME --name inference -t -d tracking configs/docker/train_baseline_data2.py
5. The directory from step 4 must contain the video and the ocr. The output of the container will be in this directory.
6. To check the process of the container, run docker log inference -f