The path to dataset can be provided using the dataset parameter and mode parameter can be used to specify the mode in which to execute perceptron. There are two modes available: erm for Empirical Risk Minimization and cv for 10 fold Cross Validation. For example:
python perceptron.py --dataset 'path/to/dataset' --mode erm