https://sites.google.com/view/alpha-mind-games/startseite/projekt-helm/einleitung

    python3 -m venv .venv
    source .venv/bin/activate

python3 -m pip -r install requirements.txt

Single .csvcan be combined via the .concatCSV.shscript located in the root directory. This script will concat all .csvfiles in the /data folder and output a file named all.csv. This file will most likely be the input parameter for the training process.

In order to predict and validate the trained model, a separated validation data file should be placed outside the /data folder to not be included in the combined .csv file.

python3 src/main.py -t data/all.csv

Preprocessed data will be saved as .npy files in the data folder. This will save a lot of computation time in successive training runs (in case mode parameters got changed). If a different data set should be used, the user has to remove those files in order to generate new .npyfiles.

python3 src/main.py -p data/toPredict.csv -m models/fullTrained1212.h5

python3 src/main.py -l -m models/fullTrained1212.h5