This code is designed to simplify the training of deep learning models for classification of videos, specifically for detecting hackers in video games.
Currently, code is implemented only in notebooks
Note that GPU passthrough is not supported on windows- and generally I only test any of this in linux (specifically ubuntu) although it should be easy to adapt to other operating systems.
You can easily setup all dependencies if you have docker installed with the included build_and_run.sh
script
cd docker
bash build_and_run.sh
This will drop you inside a docker container that has mounted the root directory of this repo.
You can then run the start_jlab.sh
script in order to start jupyter lab and open the notebooks
bash start_jlab.sh
This will start a jupyter lab server, copy the link displayed in the CLI (including the token!) to access it
Proper code is under the src
directory.
WIP code, and various experiments/examples are under the notebooks
folder