2D Pattern Classification via Linear Programming

Set Up

python -m venv env
pip install -r requirements.txt

Example Usage

python classify.py -l 2 3 -g -e 20 -f ./img/example_usage

Available Options

usage: classify.py [-h] [-g] [-r] [-f FIGURE] [-s SEED] [-n NUMBER] [-e EXTRA]
                [-p PERCENTAGE] [-x AXIS [AXIS ...]] [-l LINE [LINE ...]]
                [-d DISTANCE [DISTANCE ...]] [-c CLASSES [CLASSES ...]]

2D Pattern Classification via Linear Programming

optional arguments:
-h, --help            show this help message and exit
-g, --guides          draw guides connecting each point to the separation
                        line (default: False)
-r, --random          generate the points completely at random (default:
                        False)
-f FIGURE, --figure FIGURE
                        save the figure as a '.eps' file with the given
                        basename (default: None)
-s SEED, --seed SEED  initialize the internal state of the random number
                        generator (default: None)
-n NUMBER, --number NUMBER
                        specify the number of random points that make up the
                        training set (default: 100)
-e EXTRA, --extra EXTRA
                        specify the number of random points that make up the
                        testing set (default: 0)
-p PERCENTAGE, --percentage PERCENTAGE
                        specify the percentage of points belonging to the
                        first class (default: 0.5)
-x AXIS [AXIS ...], --axis AXIS [AXIS ...]
                        specify the lower and upper bounds of the horizontal
                        axis (default: [-25.0, 25.0])
-l LINE [LINE ...], --line LINE [LINE ...]
                        specify the slope and the y-intercept of the
                        separation line (default: [2.0, -3.0])
-d DISTANCE [DISTANCE ...], --distance DISTANCE [DISTANCE ...]
                        specify the lower and upper bounds of the distance
                        from the separation line (default: [10.0, 80.0])
-c CLASSES [CLASSES ...], --classes CLASSES [CLASSES ...]
                        specify the classes' labels (default: ['Negative',
                        'Positive'])