def test_OrdinaEncoderFlagsWrite():
h = [["Male", 1], ["Female", 3], ["Female", 2]]
hdf = pd.DataFrame(h, columns=["a", "b"])
g = Encoders("OrdinalEncoder").cacheOn()
g.train(hdf).predict(hdf)
fp: str = "tmp/df"
assert g.write(fp) == g
def test_OrdinaEncoders(X):
h = [["Male", 1], ["Female", 3], ["Female", 2]]
hdf = pd.DataFrame(h)
o = [[1, 1], [2, 3], [2, 2]]
odf = pd.DataFrame(o)
assert ((Encoders(
description_filepath="../../descriptions/pre/encoders/OHE.yaml").train(
hdf).predict(hdf) == odf).any().any())
def test_OrdinaEncoderSaveError():
h = [["Male", 1], ["Female", 3], ["Female", 2]]
hdf = pd.DataFrame(h)
g = Encoders("OrdinalEncoder")
g.train(hdf)
with pytest.raises(PasoError):
g.save()
def test_EncoderList(X):
assert Encoders("BaseNEncoder").encoders() == [
"BackwardDifferenceEncoder",
"BinaryEncoder",
"HashingEncoder",
"HelmertEncoder",
"OneHotEncoder",
"OrdinalEncoder",
"SumEncoder",
"PolynomialEncoder",
"BaseNEncoder",
"LeaveOneOutEncoder",
"TargetEncoder",
"WOEEncoder",
"MEstimateEncoder",
"JamesSteinEncoder",
"CatBoostEncoder",
"EmbeddingVectorEncoder",
]
def test_OrdinaEncoderFlagsCacheOffpredictedNot(X):
h = [["Male", 1], ["Female", 3], ["Female", 2]]
hdf = pd.DataFrame(h)
g = Encoders("OrdinalEncoder")
g.train(hdf)
assert (g.trained and not g.predicted and not g.cache) == True
def test_BaseNEncoder_no_df(X):
with pytest.raises(PasoError):
Encoders(
description_filepath="../../descriptions/pre/encoders/OHE.yaml"
).train([["Male", 1], ["Female", 3], ["Female", 2]])
def test_bad_encoder_name():
with pytest.raises(PasoError):
g = Encoders("fred")
def test_Class_init_WrongScaler():
with pytest.raises(PasoError):
g = Encoders("GORG")
def test_Class_init_NoArg():
with pytest.raises(PasoError):
g = Encoders()