def train()

in code/cv.py [0:0]

• 29 lines of code
• 2 McCabe index (conditional complexity)

def train():
    """
    Trains a Cross Validation Model with the given parameters.
    
    """
    parser = argparse.ArgumentParser()

    # Hyperparameters are described here. In this simple example we are just including one hyperparameter.
    parser.add_argument('-c', type=float, default=1.0)
    parser.add_argument('--gamma', type=float)
    parser.add_argument('--kernel', type=str)
    parser.add_argument('-k', '--k', type=int, default=5)
    parser.add_argument('--train_src', type=str)
    parser.add_argument('--test_src', type=str)
    parser.add_argument('--output_path', type=str)
    parser.add_argument('--instance_type', type=str, default="ml.c4.xlarge")
    parser.add_argument('--region', type=str, default="us-east-2")
    
    args = parser.parse_args()

    os.environ['AWS_DEFAULT_REGION'] = args.region
    sm_client = boto3.client("sagemaker")
    training_jobs = []
    # Fit k training jobs with the specified parameters.
    for f in range(args.k):
        sklearn_estimator = fit_model(instance_type=args.instance_type,
                                      output_path=args.output_path,
                                      s3_train_base_dir=args.train_src,
                                      s3_test_base_dir=args.test_src,
                                      f=f,
                                      c=args.c,
                                      gamma=args.gamma,
                                      kernel=args.kernel)
        training_jobs.append(sklearn_estimator)
        time.sleep(5) # sleeps to avoid Sagemaker Training Job API throttling

    monitor_training_jobs(training_jobs=training_jobs, sm_client=sm_client)
    score = evaluation(training_jobs=training_jobs, sm_client=sm_client)
    return score