Merge branch 'main' into feature/ohe_refactor

CHANGELOG.rst

@@ -15,12 +15,17 @@ Subsections for each version can be one of the following;
 - ``Security`` in case of vulnerabilities.
 
 Each individual change should have a link to the pull request after the description of the change.
+
 1.3.1 (unreleased)
 ------------------
 
 Changed
 ^^^^^^^
+
+- Refactored NominalToIntegerTransformer tests in new format `#261 <https://github.com/lvgig/tubular/pull/261>`_
+- Refactored GroupRareLevelsTransformer tests in new format `#259 <https://github.com/lvgig/tubular/pull/259>`_
 - DatetimeInfoExtractor.mappings_provided changed from a dict.keys() object to list so transformer is serialisable. `#258 <https://github.com/lvgig/tubular/pull/258>`_
+- Created BaseNumericTransformer class to support test refactor of numeric file
 
 1.3.0 (2024-06-13)
 ------------------

tests/base_tests.py

@@ -9,8 +9,6 @@
 import sklearn.base as b
 import test_aide as ta
 
-import tests.test_data as d
-
 
 class GenericInitTests:
     """
@@ -288,10 +286,11 @@ def test_X_non_df_error(
         self,
         initialized_transformers,
         non_df,
+        minimal_dataframe_lookup,
     ):
         """Test an error is raised if X is not passed as a pd.DataFrame."""
 
-        df = d.create_numeric_df_1()
+        df = minimal_dataframe_lookup[self.transformer_name]
 
         x = initialized_transformers[self.transformer_name]
 
@@ -306,10 +305,11 @@ def test_non_pd_type_error(
         self,
         non_series,
         initialized_transformers,
+        minimal_dataframe_lookup,
     ):
         """Test an error is raised if y is not passed as a pd.Series."""
 
-        df = d.create_numeric_df_1()
+        df = minimal_dataframe_lookup[self.transformer_name]
 
         x = initialized_transformers[self.transformer_name]
 
@@ -338,12 +338,13 @@ def test_X_no_rows_error(
     def test_Y_no_rows_error(
         self,
         initialized_transformers,
+        minimal_dataframe_lookup,
     ):
         """Test an error is raised if Y has no rows."""
 
         x = initialized_transformers[self.transformer_name]
 
-        df = pd.DataFrame({"a": 1, "b": "wow", "c": np.nan}, index=[0])
+        df = minimal_dataframe_lookup[self.transformer_name]
 
         with pytest.raises(
             ValueError,
@@ -549,10 +550,11 @@ def test_non_pd_type_error(
         self,
         non_df,
         initialized_transformers,
+        minimal_dataframe_lookup,
     ):
         """Test that an error is raised in transform is X is not a pd.DataFrame."""
 
-        df = d.create_df_8()
+        df = minimal_dataframe_lookup[self.transformer_name]
 
         x = initialized_transformers[self.transformer_name]
 
@@ -564,19 +566,20 @@ def test_non_pd_type_error(
         ):
             x_fitted.transform(X=non_df)
 
-    def test_no_rows_error(self, initialized_transformers):
+    def test_no_rows_error(self, initialized_transformers, minimal_dataframe_lookup):
         """Test an error is raised if X has no rows."""
-        df = d.create_df_8()
+
+        df = minimal_dataframe_lookup[self.transformer_name]
 
         x = initialized_transformers[self.transformer_name]
 
         x = x.fit(df, df["c"])
 
-        df = pd.DataFrame(columns=["a", "b", "c"])
+        df = df.head(0)
 
         with pytest.raises(
             ValueError,
-            match=re.escape(f"{self.transformer_name}: X has no rows; (0, 3)"),
+            match=re.escape(f"{self.transformer_name}: X has no rows; {df.shape}"),
         ):
             x.transform(df)
 
@@ -695,9 +698,10 @@ def test_columns_not_list_error(
         self,
         non_list,
         initialized_transformers,
+        minimal_dataframe_lookup,
     ):
         """Test an error is raised if self.columns is not a list."""
-        df = d.create_df_1()
+        df = minimal_dataframe_lookup[self.transformer_name]
 
         x = initialized_transformers[self.transformer_name]
 
@@ -712,9 +716,10 @@ def test_columns_not_list_error(
     def test_columns_not_in_X_error(
         self,
         initialized_transformers,
+        minimal_dataframe_lookup,
     ):
         """Test an error is raised if self.columns contains a value not in X."""
-        df = d.create_df_1()
+        df = minimal_dataframe_lookup[self.transformer_name]
 
         x = initialized_transformers[self.transformer_name]
 

tests/conftest.py

@@ -130,6 +130,9 @@ def minimal_attribute_dict():
         "BaseNominalTransformer": {
             "columns": ["b"],
         },
+        "BaseNumericTransformer": {
+            "columns": ["a", "b"],
+        },
         "BaseTransformer": {
             "columns": ["a"],
         },
@@ -233,7 +236,7 @@ def minimal_attribute_dict():
             "separator": "-",
         },
         "OrdinalEncoderTransformer": {
-            "columns": ["b"],
+            "columns": ["b"],      
         },
         "OutOfRangeNullTransformer": {
             "capping_values": {"a": [0.1, 0.2]},

tests/nominal/test_BaseNominalTransformer.py

@@ -1,5 +1,3 @@
-import re
-
 import pandas as pd
 import pytest
 from sklearn.exceptions import NotFittedError
@@ -16,7 +14,7 @@
 # The first part of this file builds out the tests for BaseNominalTransformer so that they can be
 # imported into other test files (by not starting the class name with Test)
 # The second part actually calls these tests (along with all other require tests) for the BaseNominalTransformer
-class GenericBaseNominalTransformerTests:
+class GenericNominalTransformTests(GenericTransformTests):
     """
     Tests for BaseNominalTransformer.transform().
     Note this deliberately avoids starting with "Tests" so that the tests are not run on import.
@@ -29,7 +27,7 @@ def test_not_fitted_error_raised(self, initialized_transformers):
             with pytest.raises(NotFittedError):
                 initialized_transformers[self.transformer_name].transform(df)
 
-    def test_exception_raised(self, initialized_transformers):
+    def test_non_mappable_rows_exception_raised(self, initialized_transformers):
         """Test an exception is raised if non-mappable rows are present in X."""
         df = d.create_df_1()
 
@@ -44,7 +42,7 @@ def test_exception_raised(self, initialized_transformers):
 
         with pytest.raises(
             ValueError,
-            match="BaseNominalTransformer: nulls would be introduced into column b from levels not present in mapping",
+            match=f"{self.transformer_name}: nulls would be introduced into column b from levels not present in mapping",
         ):
             x.transform(df)
 
@@ -63,24 +61,6 @@ def test_original_df_not_updated(self, initialized_transformers):
 
         pd.testing.assert_frame_equal(df, d.create_df_1())
 
-    def test_no_rows_error(self, initialized_transformers):
-        """Test an error is raised if X has no rows."""
-        df = d.create_df_1()
-
-        x = initialized_transformers[self.transformer_name]
-
-        x = x.fit(df)
-
-        x.mappings = {"b": {"a": 1, "b": 2, "c": 3, "d": 4, "e": 5, "f": 6}}
-
-        df = pd.DataFrame(columns=["a", "b", "c"])
-
-        with pytest.raises(
-            ValueError,
-            match=re.escape(f"{self.transformer_name}: X has no rows; (0, 3)"),
-        ):
-            x.transform(df)
-
 
 class TestInit(ColumnStrListInitTests):
     """Generic tests for transformer.init()."""
@@ -98,7 +78,7 @@ def setup_class(cls):
         cls.transformer_name = "BaseNominalTransformer"
 
 
-class TestTransform(GenericBaseNominalTransformerTests, GenericTransformTests):
+class TestTransform(GenericNominalTransformTests):
     """Tests for BaseImputer.transform."""
 
     @classmethod