def test_validate_np_array_with_provided_cols():
# test check_dataframe with a np.ndarray and provided cols
x, cols = check_dataframe(np.random.rand(5, 5), cols=[0, 1, 3])
assert isinstance(x, pd.DataFrame)
assert isinstance(cols, list)
assert cols == [0, 1, 3]
assert x.columns.tolist() == [0, 1, 2, 3, 4]
def transform(self, X):
"""Apply the schema normalization.
Parameters
----------
X : pd.DataFrame, shape=(n_samples, n_features)
The Pandas frame to transform. The operation will
be applied to a copy of the input data, and the result
will be returned.
Returns
-------
X : pd.DataFrame or np.ndarray, shape=(n_samples, n_features)
The operation is applied to a copy of ``X``,
and the result set is returned.
"""
check_is_fitted(self, "validator_")
X, _ = check_dataframe(X, cols=self.cols)
# make the document, normalize
v = self.validator_
X = pd.DataFrame.from_records(
[v.normalized(record) for record in X.to_dict(orient='records')])
return X if self.as_df else X.values
def test_check_dataframe_some_cols():
# a check with all columns present
X_copy, cols_copy = check_dataframe(X, cols=cols[:3])
assert X.equals(X_copy)
# cols_copy should NOT equal cols
assert cols_copy != cols
assert isinstance(cols_copy, list)
def test_check_dataframe_assert_all_finite():
# a check with all columns present
X_copy, cols_copy = check_dataframe(X, assert_all_finite=True)
assert X.equals(X_copy)
assert X_copy is not X
# X_copy should equal X
assert cols == cols_copy, (cols, cols_copy)
def test_check_dataframe_infinite():
X_nan = X.mask(X < 0.3)
# should not raise initially
X_copy, _ = check_dataframe(X_nan)
assert X_copy.equals(X_nan)
# this will raise, since assert_all_finite is True
assert_raises(ValueError, check_dataframe, X_nan, assert_all_finite=True)
def test_check_dataframe_scalar_col():
# a check with all columns present
X_copy, cols_copy = check_dataframe(X, cols='col_0')
assert X.equals(X_copy)
# cols_copy should NOT equal cols
assert cols_copy != cols
assert isinstance(cols_copy, list)
assert len(cols_copy) == 1
assert cols_copy[0] == 'col_0'
def test_check_dataframe_no_cols():
# a check with all columns present
X_copy, cols_copy = check_dataframe(X, cols=None)
assert X.equals(X_copy)
# assert cols is a list that equals ALL cols
assert cols_copy is not cols
assert cols == cols_copy, (cols, cols_copy)
assert isinstance(cols_copy, list)
assert X_copy.columns.tolist() == cols
def test_check_dataframe_with_diff():
# a check with all columns present
X_copy, cols_copy, diff = check_dataframe(X, cols=cols, column_diff=True)
# neither copy should not share the same reference (still)
assert X_copy is not X
assert cols_copy is not cols
# assert equalities
assert X.equals(X_copy)
assert cols == cols_copy, (cols, cols_copy)
assert not diff
def test_check_dataframe_all_cols():
# a check with all columns present
X_copy, cols_copy = check_dataframe(X, cols=cols)
# neither copy should not share the same reference
assert X_copy is not X
assert cols_copy is not cols
# X_copy should equal X
assert X.equals(X_copy)
assert cols == cols_copy, (cols, cols_copy)
assert isinstance(cols_copy, list)
assert X_copy.columns.tolist() == cols
def test_check_dataframe_array():
X_copy, cols_copy = check_dataframe(array, cols=None)
assert isinstance(X_copy, pd.DataFrame)
assert cols_copy == list(range(5))
