def test_plot_nulls_in_the_eda():
dataset = Dataset('src/data/Health_data.csv')
dataset.load_data()
preprocess = Preprocess(dataset.get_data())
eda = EDA(preprocess.get_preprocessed_so_far()).plot_nulls()
eda_nulls = EDA(preprocess.get_preprocessed_so_far()).plot_nulls()
assert eda == eda_nulls
def test_preprocess_delegation_to_eda_object():
'''EDA gets passed the data that has undergone preprocessing.
In the future, I can refactor to make Preprocess
immutable objects. So far EDA is immutable.'''
dataset = Dataset('src/data/Health_data.csv')
dataset.load_data()
preprocess = Preprocess(dataset.get_data())
eda = EDA(preprocess.get_preprocessed_so_far())
assert eda
def test_fill_nulls_with_zeros():
dataset = Dataset('src/data/Health_data.csv')
dataset.load_data()
preprocess = Preprocess(dataset.get_data())
# First preprocess object with filled nulls
preprocess = preprocess.fill_nulls_with_zeros()
# Another preprocess object that has filled nulls
zeros = Preprocess(dataset.get_data()).fill_nulls_with_zeros()
assert preprocess == zeros
def __init__(self, data_path, features_config, model_config):
# Composition, self contains other objects
MachineLearningApp.__init__(self)
self.Dataset = Dataset(data_path)
self.Features = Features(features_config)
self.Model = Model(model_config)
def test_right_path_and_data_dimensions():
dataset = Dataset('src/data/Health_data.csv')
assert dataset.load_data() == (5148, 19)
def test_wrong_data_path():
dataset = Dataset('/path/to/data')
assert dataset.load_data() == 'incorrect path, could not find data'
def test_delegation_to_preprocess_object():
dataset = Dataset('src/data/Health_data.csv')
preprocess = Preprocess(dataset.get_data())
assert preprocess