OpenCV 4 with Python Blueprints - Second Edition

Running the script

The MLP classifier can be trained and tested by using the train_classifier.py script, which does the following:

1. The script first sets up the command-line options of --data to the location of the saved data, and --save to the location of a directory where we want to save the trained model (this argument is optional), as follows:

if __name__ == '__main__':    parser = argparse.ArgumentParser()    parser.add_argument('--data', required=True)    parser.add_argument('--save', type=Path)    args = parser.parse_args()

2. Then, we load the saved data, and follow the training procedure described in the previous section, as follows:

    data, targets = load_collected_data(args.data)    mlp = cv2.ml.ANN_MLP_create()    ...    mlp.train(...

3. Finally, we ...