def test_robust_transformer_desugar():
"""Should be able to "desugar" multiple things into a valid transformer pipeline""" # noqa
transformer = [
None,
IdentityTransformer(),
lambda x: x,
Feature('A', IdentityTransformer()),
('A', IdentityTransformer()),
('A', [None, IdentityTransformer()]),
]
robust_transformer = make_robust_transformer(transformer)
assert isinstance(robust_transformer, TransformerPipeline)
def get_target_encoder(income_col='PINCP', income_threshold=INCOME_THRESHOLD):
"""Get encoder for the prediction target
Returns:
transformer-like
"""
# From Kaggle:
# > The prediction task is to determine whether a person makes over $50K a
# > year.
# We adjust to 2018 dollars abovee.
return make_robust_transformer([
lambda y: y[income_col],
lambda y: (y > income_threshold).astype(int),
])
def __init__(self,
input,
transformer,
name=None,
description=None,
output=None,
source=None,
options=None):
self.input = input
self.transformer = make_robust_transformer(transformer)
self.name = name
self.description = description
self.output = output # unused
self.source = source
self.options = options if options is not None else {}
def __init__(self,
input: FeatureInputType,
transformer: FeatureTransformerType,
name: Optional[str] = None,
description: Optional[str] = None,
output: Optional[OneOrMore[str]] = None,
source: Optional[str] = None,
options: Optional[dict] = None):
self.input = input
self.transformer = make_robust_transformer(transformer)
self.name = name
self.description = description
self.output = (
output
or (slugify(self.name, separator='_') if self.name else None))
self.source = source
self.options = options or {}
def test_get_transformer_primitives(transformer, expected):
robust_transformer = make_robust_transformer(transformer)
primitives = get_transformer_primitives(robust_transformer)
assert primitives == expected
sklearn.preprocessing.StandardScaler,
sklearn.preprocessing.Binarizer,
sklearn.preprocessing.PolynomialFeatures,
)
# some of these input types are bad for sklearn.
input_types = ('ser', 'df', 'arr1d')
for Transformer in Transformers:
robust_transformer = DelegatingRobustTransformer(Transformer())
for input_type in input_types:
X, y = sample_data[input_type]
robust_transformer.fit_transform(X, y=y)
@pytest.mark.parametrize('robust_maker', [
DelegatingRobustTransformer,
lambda x: make_robust_transformer([x]),
])
def test_robust_str_repr(robust_maker):
robust_transformer = robust_maker(IdentityTransformer())
for func in [str, repr]:
s = func(robust_transformer)
assert len(s) > 0
@pytest.mark.parametrize(
'transformer,expected',
[(
IdentityTransformer(),
['IdentityTransformer'],
),
(