Remove remote dataset retrieval for tests

.gitignore

@@ -24,6 +24,7 @@ wheels/
 .installed.cfg
 *.egg
 MANIFEST
+.uv.lock
 
 # PyInstaller
 #  Usually these files are written by a python script from a template
@@ -111,4 +112,4 @@ venv.bak/
 *.csv
 *.DS_Store
 *.db
-*.pptx
+*.pptx

tests/conftest.py

@@ -1,3 +1,5 @@
+from pathlib import Path
+
 import numpy as np
 import pandas as pd
 import pytest
@@ -199,3 +201,10 @@ def df_normal_dist():
     df.columns = ["var"]
 
     return df
+
+
+@pytest.fixture(scope="module")
+def static_california_housing() -> pd.DataFrame:
+    return pd.read_csv(
+        Path(__file__).parent / "resources" / "california_housing.csv", index_col=0
+    )

tests/test_discretisation/test_arbitrary_discretiser.py

@@ -3,16 +3,12 @@
 import pytest
 from numpy.random import default_rng
 from scipy.stats import skewnorm
-from sklearn.datasets import fetch_california_housing
 
 from feature_engine.discretisation import ArbitraryDiscretiser
 
 
-def test_arbitrary_discretiser():
-    california_dataset = fetch_california_housing()
-    data = pd.DataFrame(
-        california_dataset.data, columns=california_dataset.feature_names
-    )
+def test_arbitrary_discretiser(static_california_housing):
+    data = static_california_housing
     user_dict = {"HouseAge": [0, 20, 40, 60, np.inf]}
 
     data_t1 = data.copy()

tests/test_discretisation/test_base_discretizer.py

@@ -1,7 +1,6 @@
 import numpy as np
 import pandas as pd
 import pytest
-from sklearn.datasets import fetch_california_housing
 
 from feature_engine.discretisation.base_discretiser import BaseDiscretiser
 
@@ -37,23 +36,21 @@ def test_correct_param_assignment_at_init(params):
 
 
 class MockClassFit(BaseDiscretiser):
+    def __init__(self, dataset: pd.DataFrame, *args, **kwargs):
+        self._dataset = dataset
+        super().__init__(*args, **kwargs)
+
     def fit(self, X):
-        california_dataset = fetch_california_housing()
-        data = pd.DataFrame(
-            california_dataset.data, columns=california_dataset.feature_names
-        )
+        data = self._dataset
         self.variables_ = ["HouseAge"]
         self.binner_dict_ = {"HouseAge": [0, 20, 40, 60, np.inf]}
         self.n_features_in_ = data.shape[1]
-        self.feature_names_in_ = california_dataset.feature_names
+        self.feature_names_in_ = data.columns
         return self
 
 
-def test_transform():
-    california_dataset = fetch_california_housing()
-    data = pd.DataFrame(
-        california_dataset.data, columns=california_dataset.feature_names
-    )
+def test_transform(static_california_housing):
+    data = static_california_housing
 
     data_t1 = data.copy()
     data_t2 = data.copy()
@@ -70,10 +67,10 @@ def test_transform():
         include_lowest=True,
     )
 
-    transformer = MockClassFit(return_boundaries=False)
+    transformer = MockClassFit(data, return_boundaries=False)
     X = transformer.fit_transform(data)
     pd.testing.assert_frame_equal(X, data_t2)
 
-    transformer = MockClassFit(return_object=False, return_boundaries=True)
+    transformer = MockClassFit(data, return_object=False, return_boundaries=True)
     X = transformer.fit_transform(data)
     pd.testing.assert_frame_equal(X, data_t1)

tests/test_selection/test_mrmr.py

@@ -5,7 +5,6 @@
 from sklearn.datasets import (
     make_classification,
     load_breast_cancer,
-    fetch_california_housing,
 )
 from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor
 from sklearn.feature_selection import (
@@ -300,8 +299,9 @@ def test_mrmr_mid_and_miq_classif():
     pd.testing.assert_frame_equal(X.drop(to_drop, axis=1), Xtr)
 
 
-def test_mrmr_fcd_and_fcq_regression():
-    X, y = fetch_california_housing(return_X_y=True, as_frame=True)
+def test_mrmr_fcd_and_fcq_regression(static_california_housing):
+    X = static_california_housing
+    y = X["MedHouseVal"]
 
     selected = ["MedInc", "Latitude", "HouseAge", "AveRooms", "AveOccup"]
     to_drop = [f for f in X.columns if f not in selected]

tests/test_wrappers/test_sklearn_wrapper.py

@@ -3,7 +3,6 @@
 import pytest
 from sklearn import __version__ as skl_version
 from sklearn.base import clone
-from sklearn.datasets import fetch_california_housing
 from sklearn.decomposition import PCA
 from sklearn.ensemble import RandomForestClassifier, VotingClassifier
 from sklearn.feature_selection import (
@@ -32,7 +31,6 @@
 
 from feature_engine.wrappers import SklearnTransformerWrapper
 
-
 if skl_version < "1.7.0":
     kbd = KBinsDiscretizer(n_bins=3, encode="ordinal")
 else:
@@ -112,9 +110,9 @@ def test_error_not_implemented_transformer(transformer, df_na):
 
 
 @pytest.mark.parametrize("transformer", _selectors)
-def test_wrap_selectors(transformer):
+def test_wrap_selectors(transformer, static_california_housing):
     # load data
-    X = fetch_california_housing(as_frame=True).frame
+    X = static_california_housing
     y = X["MedHouseVal"]
     X = X.drop(["MedHouseVal"], axis=1)
 
@@ -150,9 +148,9 @@ def test_wrap_selectors(transformer):
 
 
 @pytest.mark.parametrize("transformer", _transformers)
-def test_wrap_transformers(transformer):
+def test_wrap_transformers(transformer, static_california_housing):
     # load data
-    X = fetch_california_housing(as_frame=True).frame
+    X = static_california_housing
 
     # prepare selectors
     tr = clone(transformer)
@@ -182,9 +180,9 @@ def test_wrap_transformers(transformer):
     pd.testing.assert_frame_equal(Xt, Xw)
 
 
-def test_wrap_polynomial_features():
+def test_wrap_polynomial_features(static_california_housing):
     # load data
-    X = fetch_california_housing(as_frame=True).frame
+    X = static_california_housing
 
     # prepare selectors
     tr = PolynomialFeatures()
@@ -215,8 +213,8 @@ def test_wrap_polynomial_features():
     assert Xw.shape[1] == len(tr.get_feature_names_out(X.columns))
 
 
-def test_wrap_polynomial_features_get_features_name_out():
-    X = fetch_california_housing(as_frame=True).frame
+def test_wrap_polynomial_features_get_features_name_out(static_california_housing):
+    X = static_california_housing
 
     varlist = ["MedInc", "HouseAge", "AveRooms", "AveBedrms"]
     tr_wrap = SklearnTransformerWrapper(
@@ -521,14 +519,14 @@ def test_sklearn_ohe_all_features(df_vartypes):
     pd.testing.assert_frame_equal(ref, transformed_df)
 
 
-def test_sklearn_ohe_with_crossvalidation():
+def test_sklearn_ohe_with_crossvalidation(static_california_housing):
     """
     Created 2022-02-14 to test fix to issue # 368
     """
 
     # Set up test pipeline with wrapped OneHotEncoder, with simple regression model
     # to be able to run cross-validation; use sklearn CA housing data
-    df = fetch_california_housing(as_frame=True).frame
+    df = static_california_housing
     y = df["MedHouseVal"]
     X = (
         df[["HouseAge", "AveBedrms"]]
@@ -614,8 +612,8 @@ def test_wrap_one_hot_encoder_get_features_name_out(df_vartypes):
     "transformer",
     [PowerTransformer(), OrdinalEncoder(), MinMaxScaler(), StandardScaler()],
 )
-def test_inverse_transform(transformer):
-    X = fetch_california_housing(as_frame=True).frame
+def test_inverse_transform(transformer, static_california_housing):
+    X = static_california_housing
     X = X.drop(["Longitude"], axis=1)
 
     tr_wrap = SklearnTransformerWrapper(transformer=transformer)
@@ -645,8 +643,10 @@ def test_inverse_transform(transformer):
         SimpleImputer(),
     ],
 )
-def test_error_when_inverse_transform_not_implemented(transformer):
-    X = fetch_california_housing(as_frame=True).frame
+def test_error_when_inverse_transform_not_implemented(
+    transformer, static_california_housing
+):
+    X = static_california_housing
     y = X["MedHouseVal"]
     X = X.drop(["MedHouseVal"], axis=1)
 
@@ -662,8 +662,10 @@ def test_error_when_inverse_transform_not_implemented(transformer):
     "varlist", [["MedInc", "HouseAge", "AveRooms", "AveBedrms"], None]
 )
 @pytest.mark.parametrize("transformer", _transformers)
-def test_get_feature_names_out_transformers(varlist, transformer):
-    X = fetch_california_housing(as_frame=True).frame
+def test_get_feature_names_out_transformers(
+    varlist, transformer, static_california_housing
+):
+    X = static_california_housing
     tr_wrap = SklearnTransformerWrapper(transformer=transformer, variables=varlist)
     Xw = tr_wrap.fit_transform(X)
 
@@ -675,8 +677,10 @@ def test_get_feature_names_out_transformers(varlist, transformer):
     "varlist", [["MedInc", "HouseAge", "AveRooms", "AveBedrms"], None]
 )
 @pytest.mark.parametrize("transformer", _selectors)
-def test_get_feature_names_out_selectors(varlist, transformer):
-    X = fetch_california_housing(as_frame=True).frame
+def test_get_feature_names_out_selectors(
+    varlist, transformer, static_california_housing
+):
+    X = static_california_housing
     y = X["MedHouseVal"]
     X = X.drop(["MedHouseVal"], axis=1)
     tr_wrap = SklearnTransformerWrapper(transformer=transformer, variables=varlist)
@@ -689,8 +693,8 @@ def test_get_feature_names_out_selectors(varlist, transformer):
 @pytest.mark.parametrize(
     "varlist", [["MedInc", "HouseAge", "AveRooms", "AveBedrms"], None]
 )
-def test_get_feature_names_out_polynomialfeatures(varlist):
-    X = fetch_california_housing(as_frame=True).frame
+def test_get_feature_names_out_polynomialfeatures(varlist, static_california_housing):
+    X = static_california_housing
     tr_wrap = SklearnTransformerWrapper(
         transformer=PolynomialFeatures(), variables=varlist
     )