def test_df_Lambert_wo(df_City):
g = Scaler("LambertScaler").cacheOn()
g.train(df_City).predict(df_City)
fp: str = "tmp/df_write"
g.write(fp)
assert (g.trained and g.predicted and g.cache and g.persisted
and (g.save_file_name == fp)) == True
def test_df_MinMax_write(df_type):
g = Scaler("MinMaxScaler").cacheOn()
g.train(df_type).predict(df_type)
fp: str = "tmp/df"
g.write(fp)
assert (g.trained and g.predicted and g.cache and g.persisted
and (g.save_file_name == fp)) == True
def test_df_BoxCox_read(df_type):
g = Scaler("BoxCoxScaler").cacheOn()
g.train(df_type).predict(df_type)
fp: str = "tmp/df_write"
g.write(fp)
g.read(fp)
assert (g.trained and g.predicted and g.cache and g.persisted
and (g.save_file_name == fp)) == True
def test_df_Lambert_type_error(ystr):
g = Scaler("LambertScaler")
with pytest.raises(PasoError):
g.train(ystr)
def test_df_Lambert_numpy_1d_error(y):
g = Scaler("LambertScaler")
with pytest.raises(PasoError):
g.train(y)
def test_df_Lambert_predict(df_type):
g = Scaler("LambertScaler")
g.train(df_type)
assert g.predict(df_type).shape == df_type.shape
def test_df_Lambert_train(df_type):
g = Scaler("LambertScaler")
assert g.train(df_type) == g
def test_df_BoxCox_type_error(ystr):
g = Scaler("BoxCoxScaler")
with pytest.raises(TypeError):
g.train(ystr())
def test_predict_df_BoxCox_df_type(df_type):
g = Scaler("BoxCoxScaler")
g.train(df_type)
assert g.predict(df_type).shape == df_type.shape
def test_df_BoxCox_numpy_1d_error(y):
g = Scaler("BoxCoxScaler")
with pytest.raises(PasoError):
g.train(y)
def test_df_BoxCox_df_type(df_type):
g = Scaler("BoxCoxScaler")
assert g.train(df_type) == g
def test_df_BoxCox_City_NA_error(df_small):
g = Scaler("BoxCoxScaler")
with pytest.raises(PasoError):
g.train(df_small)
def test_df_BoxCox_City_negative_error(df_City):
g = Scaler("BoxCoxScaler")
with pytest.raises(ValueError):
g.train(df_City)
def test_df_BoxCox_inverse(df_type):
g = Scaler("BoxCoxScaler")
g.train(df_type, inplace=False)
assert g.inverse_predict(g.predict(df_type) == df_type).any().any()