Skip to content

add validation for ignore_attributes and default_target_attribute at … #978

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
PGijsbers merged 7 commits into from
Oct 29, 2020
Merged

add validation for ignore_attributes and default_target_attribute at … #978

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
1c6c5de
add validation for ignore_attributes and default_target_attribute at …
madel0093 Oct 28, 2020
07c85cc
update naming convetions and adding type hints. using pytest parametr…
madel0093 Oct 28, 2020
ff3d27a
formating long lines and update types hint for return values
madel0093 Oct 28, 2020
ace60aa
update test_attribute_validations to use pytest.mark.parametrize
madel0093 Oct 29, 2020
afdd949
add more tests for different input types for attribute validation
madel0093 Oct 29, 2020
4cf640f
update formatting
madel0093 Oct 29, 2020
89c2ada
Merge branch 'develop' into feature/validate_attributes_at_create_dat…
a-moadel Oct 29, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 27 additions & 0 deletions openml/datasets/functions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -333,6 +333,27 @@ def _load_features_from_file(features_file: str) -> Dict:
return xml_dict["oml:data_features"]


def _expand_parameter(parameter: Union[str, list]):
Comment thread
PGijsbers marked this conversation as resolved.
Outdated
expanded_parameter = []
if isinstance(parameter, str):
expanded_parameter = [x.strip() for x in parameter.split(",")]
elif isinstance(parameter, list):
expanded_parameter = parameter
return expanded_parameter


def _validated_data_attributes(attributes: list, data_attributes: list, parameter_name: str):
Comment thread
PGijsbers marked this conversation as resolved.
Outdated
for attribute_ in attributes:
is_attribute_a_data_attribute = any([attr[0] == attribute_ for attr in data_attributes])
if not is_attribute_a_data_attribute:
raise ValueError(
"all attribute of '{}' should be one of the data attribute. "
" Got '{}' while candidates are {}.".format(
parameter_name, attribute_, [attr[0] for attr in data_attributes]
)
)


def check_datasets_active(dataset_ids: List[int]) -> Dict[int, bool]:
"""
Check if the dataset ids provided are active.
Expand Down Expand Up @@ -636,6 +657,7 @@ def create_dataset(
ignore_attribute : str | list
Attributes that should be excluded in modelling,
such as identifiers and indexes.
Can have multiple values, comma separated.
citation : str
Reference(s) that should be cited when building on this data.
version_label : str, optional
Expand Down Expand Up @@ -687,6 +709,11 @@ def create_dataset(
attributes_[attr_idx] = (attr_name, attributes[attr_name])
else:
attributes_ = attributes
ignore_attributes = _expand_parameter(ignore_attribute)
_validated_data_attributes(ignore_attributes, attributes_, "ignore_attribute")

default_target_attributes = _expand_parameter(default_target_attribute)
_validated_data_attributes(default_target_attributes, attributes_, "default_target_attribute")

if row_id_attribute is not None:
is_row_id_an_attribute = any([attr[0] == row_id_attribute for attr in attributes_])
Expand Down
52 changes: 48 additions & 4 deletions tests/test_datasets/test_dataset_functions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -875,6 +875,51 @@ def test_get_online_dataset_format(self):
_get_online_dataset_format(dataset_id),
"The format of the ARFF files is different",
)
@pytest.mark.parametrize("default_target_attribute,row_id_attribute,ignore_attribute",
[("wrong", None,None), (None,"wrong",None), (None,None,"wrong")])
def test_attribute_validations(self):
data = [
["a", "sunny", 85.0, 85.0, "FALSE", "no"],
["b", "sunny", 80.0, 90.0, "TRUE", "no"],
["c", "overcast", 83.0, 86.0, "FALSE", "yes"],
["d", "rainy", 70.0, 96.0, "FALSE", "yes"],
["e", "rainy", 68.0, 80.0, "FALSE", "yes"],
]
column_names = ["rnd_str", "outlook", "temperature", "humidity", "windy", "play"]
df = pd.DataFrame(data, columns=column_names)
# enforce the type of each column
df["outlook"] = df["outlook"].astype("category")
df["windy"] = df["windy"].astype("bool")
df["play"] = df["play"].astype("category")
# meta-information
name = "%s-pandas_testing_dataset" % self._get_sentinel()
description = "Synthetic dataset created from a Pandas DataFrame"
creator = "OpenML tester"
collection_date = "01-01-2018"
language = "English"
licence = "MIT"
citation = "None"
original_data_url = "http://openml.github.io/openml-python"
paper_url = "http://openml.github.io/openml-python"
with self.assertRaises(ValueError):
_ = openml.datasets.functions.create_dataset(
name=name,
description=description,
creator=creator,
contributor=None,
collection_date=collection_date,
language=language,
licence=licence,
default_target_attribute="wrong",
row_id_attribute=None,
ignore_attribute=None,
citation=citation,
attributes="auto",
data=df,
version_label="test",
original_data_url=original_data_url,
paper_url=paper_url,
)

def test_create_dataset_pandas(self):
data = [
Expand All @@ -897,7 +942,6 @@ def test_create_dataset_pandas(self):
collection_date = "01-01-2018"
language = "English"
licence = "MIT"
default_target_attribute = "play"
citation = "None"
original_data_url = "http://openml.github.io/openml-python"
paper_url = "http://openml.github.io/openml-python"
Expand All @@ -909,7 +953,7 @@ def test_create_dataset_pandas(self):
collection_date=collection_date,
language=language,
licence=licence,
default_target_attribute=default_target_attribute,
default_target_attribute="play",
Copy link
Copy Markdown
Collaborator

@PGijsbers PGijsbers Oct 28, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why did you change these tests?

Copy link
Copy Markdown
Contributor Author

@a-moadel a-moadel Oct 28, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This one will fail because of the new logic , as "play" is not in the dataset columns.

Copy link
Copy Markdown
Collaborator

@PGijsbers PGijsbers Oct 28, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

But previously it was also set to "play"? And it says column_names = ["rnd_str", "outlook", "temperature", "humidity", "windy", "play"] just a few lines up. I am a bit confused about this. Maybe sitting together for this works better.

row_id_attribute=None,
ignore_attribute=None,
citation=citation,
Expand Down Expand Up @@ -944,7 +988,7 @@ def test_create_dataset_pandas(self):
collection_date=collection_date,
language=language,
licence=licence,
default_target_attribute=default_target_attribute,
default_target_attribute="y",
row_id_attribute=None,
ignore_attribute=None,
citation=citation,
Expand Down Expand Up @@ -980,7 +1024,7 @@ def test_create_dataset_pandas(self):
collection_date=collection_date,
language=language,
licence=licence,
default_target_attribute=default_target_attribute,
default_target_attribute="rnd_str",
row_id_attribute=None,
ignore_attribute=None,
citation=citation,
Expand Down