def fit_model()

in code/cv.py [0:0]

• 27 lines of code
• 1 McCabe index (conditional complexity)

def fit_model(instance_type, 
              output_path,
              s3_train_base_dir,
              s3_test_base_dir,
              f, 
              c, 
              gamma, 
              kernel
             ):
    """Fits a model using the specified algorithm.
    
       Args:
            instance_type: instance to use for Sagemaker Training job
            output_path: S3 URI as the location for the trained model artifact
            s3_train_base_dir: S3 URI for train datasets
            s3_test_base_dir: S3 URI for test datasets
            f: index represents a fold number in the K fold cross validation
            c: regularization parameter for SVM
            gamma: kernel coefficiency value
            kernel: kernel type for SVM algorithm
  
       
       Returns: 
            Sagemaker Estimator created with given input parameters.
    """
    sklearn_framework_version='0.23-1'
    script_path = 'scikit_learn_iris.py'

    sagemaker_session = sagemaker.Session()
    role = sagemaker.get_execution_role()

    sklearn_estimator = SKLearn(
        entry_point=script_path,
        instance_type=instance_type,
        framework_version=sklearn_framework_version,
        role=role,
        sagemaker_session=sagemaker_session,
        output_path=output_path,
        hyperparameters={'c': c, 'gamma' : gamma, 'kernel': kernel},
        metric_definitions= [ { "Name": "test:score", "Regex": "model test score:(.*?);" }]
    )
    sklearn_estimator.fit(inputs = { 'train': f'{s3_train_base_dir}/{f}',
                                     'test':  f'{s3_test_base_dir}/{f}'
                                   }, wait=False)
    return sklearn_estimator