def test_xtrain_larger_than_xtest(): # given test_xtrain_data = load_dataset(file_name='xtrain.csv') test_xtrain_data_size = len(test_xtrain_data.index) # when test_xtest_data = load_dataset(file_name='xtest.csv') test_xtest_data_size = len(test_xtest_data.index) # then assert test_xtrain_data_size is not None assert test_xtest_data_size is not None assert test_xtrain_data_size > test_xtest_data_size
def test_make_single_prediction():
test_data = load_dataset(file_name='test.csv')
single_test_input = test_data[0:1]
subject = make_prediction(input_data=single_test_input)
assert subject is not None
assert isinstance(subject.get('prediction')[0], float)
assert math.ceil(subject.get('prediction')[0]) == 112476
def test_xtest_playerNname_dtype():
# given
test_data_xtest = load_dataset(file_name='xtest.csv')
# when
correct_dtype = np.dtype(np.int64)
test_dtype_xtest = test_data_xtest["playerName"].dtypes
# then
assert test_dtype_xtest == correct_dtype
def test_xtrain_column_length(): # given test_data = load_dataset(file_name='xtrain.csv') test_data_column_len = len(test_data.columns) - 6 # when correct_column_len = len(cfg.FEATURE_LIST) # then assert test_data_column_len is not None assert test_data_column_len == correct_column_len
def test_prediction_endpoint_returns_prediction(flask_test_client):
test_data = load_dataset(file_name=model_config.TESTING_DATA_FILE)
post_json = test_data[0:1].to_json(orient='records')
response = flask_test_client.post('/v1/predict/regression', json=post_json)
assert response.status_code == 200
response_json = json.loads(response.data)
prediction = response_json['predictions']
response_version = response_json['version']
assert math.ceil(prediction) == 112476
assert response_version == _version
def test_prediction_endpoint_validation_200(flask_test_client):
test_data = load_dataset(file_name=config.TESTING_DATA_FILE)
post_json = test_data.to_json(orient='records')
response = flask_test_client.post('/v1/predict/regression',
json=post_json)
assert response.status_code == 200
response_json = json.loads(response.data)
assert len(response_json.get('predictions')) + len(
response_json.get('errors')) == len(test_data)
def test_make_multiple_predictions():
test_data = load_dataset(file_name='test.csv')
original_data_length = len(test_data)
multiple_test_input = test_data
subject = make_prediction(input_data=multiple_test_input)
assert subject is not None
assert len(subject.get('prediction')) == 1451
assert len(subject.get('prediction')) != original_data_length
def test_make_single_prediction():
# given
test_data = load_dataset(file_name='xtest.csv')
single_test_json = test_data[0:1].to_json(orient='records')
# reads the 1st line of 'test.csv'
# when
subject = make_prediction(input_data=single_test_json)
# print(subject.get('predictions')[0])
# the above print statement was used to get the correct predicted value for
# line 22 ~ subject to change after editing
# then
assert subject is not None
assert isinstance(subject.get('predictions')[0],
float) # calling the 'predictions' keyword from the dict
assert math.ceil(subject.get('predictions')[0]) == 16 # see line 17
def test_make_multiple_predictions():
# given
test_data = load_dataset(file_name='xtest.csv')
original_data_length = len(test_data)
multiple_test_json = test_data.to_json(orient='records')
# when
subject = make_prediction(input_data=multiple_test_json)
# print(original_data_length) # only work when tests fail
# print(subject.get('predictions')) --> verified the points for each individual player is outputted
# then
assert subject is not None
assert len(subject.get('predictions')) == 1034 # white box testing
# we dont expect any rows to be filtered out
assert len(subject.get('predictions')) == original_data_length
def run_training() -> None:
""" Train the model"""
# read training data
data = load_dataset(file_name=config.TRAINING_DATA_FILE)
# divide train and test
X_train, X_test, y_train, y_test = train_test_split(
data[config.FEATURES],
data[config.TARGET],
test_size=0.1,
random_state=0) # Setting the seed
# transform target
y_train = np.log(y_train)
# y_test = np.log(y_test)
pipeline.price_pipe.fit(X_train[config.FEATURES], y_train)
_logger.info(f"saving model version: {_version}")
save_pipeline(pipeline_to_persist=pipeline.price_pipe)
def run_training():
print('Training the model...')
# read training data
data = load_dataset(file_name=cfg.TRAINING_DATA_FILE)
# divide train and test
X_train, X_test, y_train, y_test = train_test_split(
data[cfg.FEATURE_LIST],
data[cfg.TARGET],
test_size=0.2,
random_state=0) # setting the seed here
# if data tranformed then the target would be transformed
# here as well
pipeline.ffml_pipe.fit(X_train[cfg.FEATURE_LIST], y_train)
_logger.info(f"saving model version: {_version}")
save_pipeline(pipeline_to_persist=pipeline.ffml_pipe)
print('Model trained...')
