feat: better names for properties of TaggedTable (#74)

### Summary of Changes In `TaggedTable`: * Rename `feature_vectors` to `features` * Rename `target_values` to `target` --------- Co-authored-by: lars-reimann <[email protected]>
Safe-DS · Mar 24, 2023 · fee398b · fee398b
1 parent 5d5f5a6
commit fee398b
Show file tree

Hide file tree

Showing 37 changed files with 90 additions and 92 deletions.
diff --git a/docs/tutorials/machine_learning.ipynb b/docs/tutorials/machine_learning.ipynb
@@ -34,7 +34,7 @@
     "\n",
     "tagged_table = TaggedTable(\n",
     "    training_set,\n",
-    "    target_column=\"result\"\n",
+    "    target_name=\"result\"\n",
     ")"
    ],
    "metadata": {

diff --git a/src/safeds/data/tabular/containers/_tagged_table.py b/src/safeds/data/tabular/containers/_tagged_table.py
@@ -11,23 +11,23 @@ class TaggedTable(Table):
     Parameters
     ----------
     table : Table
-        The table used to derive the feature and target vectors.
-    target_column : str
+        The table used to derive the features and target.
+    target_name : str
         Name of the target column.
     """
 
-    def __init__(self, table: Table, target_column: str):
+    def __init__(self, table: Table, target_name: str):
         super().__init__(table._data)
 
-        self._y: Column = table.get_column(target_column)
-        self._X: Table = table.drop_columns([target_column])
+        self._y: Column = table.get_column(target_name)
+        self._X: Table = table.drop_columns([target_name])
 
     @property
-    def feature_vectors(self) -> Table:
+    def features(self) -> Table:
         return self._X
 
     @property
-    def target_values(self) -> Column:
+    def target(self) -> Column:
         return self._y
 
     def __repr__(self) -> str:

diff --git a/src/safeds/ml/_util_sklearn.py b/src/safeds/ml/_util_sklearn.py
@@ -24,8 +24,8 @@ def fit(model: Any, tagged_table: TaggedTable) -> None:
     """
     try:
         model.fit(
-            tagged_table.feature_vectors._data,
-            tagged_table.target_values._data,
+            tagged_table.features._data,
+            tagged_table.target._data,
         )
     except ValueError as exception:
         raise LearningError(str(exception)) from exception
@@ -41,16 +41,16 @@ def predict(model: Any, dataset: Table, target_name: str) -> TaggedTable:
     Parameters
     ----------
     model
-        Classifier or Regression from scikit-learn.
+        Classifier or regressor from scikit-learn.
     dataset : Table
-        The dataset containing the feature vectors.
+        The dataset containing the features.
     target_name : str
-        The name of the target vector, the name of the target column inferred from fit is used by default.
+        The name of the target column.
 
     Returns
     -------
     table : Table
-        A dataset containing the given feature vectors and the predicted target vector.
+        A dataset containing the given features and the predicted target.
 
     Raises
     ------
@@ -67,7 +67,7 @@ def predict(model: Any, dataset: Table, target_name: str) -> TaggedTable:
                 f"Dataset already contains '{target_name}' column. Please rename this column"
             )
         result_set[target_name] = predicted_target_vector
-        return TaggedTable(Table(result_set), target_column=target_name)
+        return TaggedTable(Table(result_set), target_name=target_name)
     except NotFittedError as exception:
         raise PredictionError("The model was not trained") from exception
     except ValueError as exception:

diff --git a/src/safeds/ml/classification/_ada_boost.py b/src/safeds/ml/classification/_ada_boost.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_classifier, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/classification/_decision_tree.py b/src/safeds/ml/classification/_decision_tree.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_classifier, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/classification/_gradient_boosting_classification.py b/src/safeds/ml/classification/_gradient_boosting_classification.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_classifier, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     # noinspection PyProtectedMember
     def predict(self, dataset: Table) -> TaggedTable:

diff --git a/src/safeds/ml/classification/_k_nearest_neighbors.py b/src/safeds/ml/classification/_k_nearest_neighbors.py
@@ -38,7 +38,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_classifier, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/classification/_logistic_regression.py b/src/safeds/ml/classification/_logistic_regression.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_classifier, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/classification/_random_forest.py b/src/safeds/ml/classification/_random_forest.py
@@ -31,7 +31,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_classifier, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_ada_boost.py b/src/safeds/ml/regression/_ada_boost.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_decision_tree.py b/src/safeds/ml/regression/_decision_tree.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_elastic_net_regression.py b/src/safeds/ml/regression/_elastic_net_regression.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_gradient_boosting_regression.py b/src/safeds/ml/regression/_gradient_boosting_regression.py
@@ -33,7 +33,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
         # noinspection PyProtectedMember
 

diff --git a/src/safeds/ml/regression/_k_nearest_neighbors.py b/src/safeds/ml/regression/_k_nearest_neighbors.py
@@ -36,7 +36,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_lasso_regression.py b/src/safeds/ml/regression/_lasso_regression.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_linear_regression.py b/src/safeds/ml/regression/_linear_regression.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_random_forest.py b/src/safeds/ml/regression/_random_forest.py
@@ -31,7 +31,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/src/safeds/ml/regression/_ridge_regression.py b/src/safeds/ml/regression/_ridge_regression.py
@@ -32,7 +32,7 @@ def fit(self, training_set: TaggedTable) -> None:
             If the tagged table contains invalid values or if the training failed.
         """
         fit(self._wrapped_regressor, training_set)
-        self._target_name = training_set.target_values.name
+        self._target_name = training_set.target.name
 
     def predict(self, dataset: Table) -> TaggedTable:
         """

diff --git a/tests/safeds/data/tabular/containers/_tagged_table/test_feature_vectors.py b/tests/safeds/data/tabular/containers/_tagged_table/test_feature_vectors.py
diff --git a/tests/safeds/data/tabular/containers/_tagged_table/test_features.py b/tests/safeds/data/tabular/containers/_tagged_table/test_features.py
@@ -0,0 +1,11 @@
+from safeds.data.tabular.containers import Table, TaggedTable
+from tests.fixtures import resolve_resource_path
+
+
+def test_tagged_table_features() -> None:
+    table = Table.from_csv(resolve_resource_path("test_tagged_table.csv"))
+    tagged_table = TaggedTable(table, "T")
+    assert "T" not in tagged_table.features._data
+    assert tagged_table.features.schema.has_column("A")
+    assert tagged_table.features.schema.has_column("B")
+    assert tagged_table.features.schema.has_column("C")
diff --git a/...iners/_tagged_table/test_target_values.py → ...r/containers/_tagged_table/test_target.py b/...iners/_tagged_table/test_target_values.py → ...r/containers/_tagged_table/test_target.py
@@ -2,8 +2,8 @@
 from tests.fixtures import resolve_resource_path
 
 
-def test_tagged_table_target_values() -> None:
+def test_tagged_table_target() -> None:
     table = Table.from_csv(resolve_resource_path("test_tagged_table.csv"))
     tagged_table = TaggedTable(table, "T")
-    assert tagged_table.target_values._data[0] == 0
-    assert tagged_table.target_values._data[1] == 1
+    assert tagged_table.target._data[0] == 0
+    assert tagged_table.target._data[1] == 1
diff --git a/tests/safeds/ml/classification/_ada_boost/test_predict.py b/tests/safeds/ml/classification/_ada_boost/test_predict.py
@@ -10,7 +10,7 @@ def test_ada_boost_predict() -> None:
     tagged_table = TaggedTable(table, "T")
     ada_boost = AdaBoost()
     ada_boost.fit(tagged_table)
-    ada_boost.predict(tagged_table.feature_vectors)
+    ada_boost.predict(tagged_table.features)
     assert True  # This asserts that the predict method succeeds
 
 
@@ -19,7 +19,7 @@ def test_ada_boost_predict_not_fitted() -> None:
     tagged_table = TaggedTable(table, "T")
     ada_boost = AdaBoost()
     with pytest.raises(PredictionError):
-        ada_boost.predict(tagged_table.feature_vectors)
+        ada_boost.predict(tagged_table.features)
 
 
 def test_ada_boost_predict_invalid() -> None:
@@ -30,4 +30,4 @@ def test_ada_boost_predict_invalid() -> None:
     ada_boost = AdaBoost()
     ada_boost.fit(tagged_table)
     with pytest.raises(PredictionError):
-        ada_boost.predict(invalid_tagged_table.feature_vectors)
+        ada_boost.predict(invalid_tagged_table.features)
diff --git a/tests/safeds/ml/classification/_decision_tree/test_predict.py b/tests/safeds/ml/classification/_decision_tree/test_predict.py
@@ -10,7 +10,7 @@ def test_decision_tree_predict() -> None:
     tagged_table = TaggedTable(table, "T")
     decision_tree = DecisionTree()
     decision_tree.fit(tagged_table)
-    decision_tree.predict(tagged_table.feature_vectors)
+    decision_tree.predict(tagged_table.features)
     assert True  # This asserts that the predict method succeeds
 
 
@@ -19,7 +19,7 @@ def test_decision_tree_predict_not_fitted() -> None:
     tagged_table = TaggedTable(table, "T")
     decision_tree = DecisionTree()
     with pytest.raises(PredictionError):
-        decision_tree.predict(tagged_table.feature_vectors)
+        decision_tree.predict(tagged_table.features)
 
 
 def test_decision_tree_predict_invalid() -> None:
@@ -32,4 +32,4 @@ def test_decision_tree_predict_invalid() -> None:
     decision_tree = DecisionTree()
     decision_tree.fit(tagged_table)
     with pytest.raises(PredictionError):
-        decision_tree.predict(invalid_tagged_table.feature_vectors)
+        decision_tree.predict(invalid_tagged_table.features)
diff --git a/tests/safeds/ml/classification/_gradient_boosting/test_predict.py b/tests/safeds/ml/classification/_gradient_boosting/test_predict.py
@@ -12,7 +12,7 @@ def test_gradient_boosting_predict() -> None:
     tagged_table = TaggedTable(table, "T")
     gradient_boosting_classification = GradientBoosting()
     gradient_boosting_classification.fit(tagged_table)
-    gradient_boosting_classification.predict(tagged_table.feature_vectors)
+    gradient_boosting_classification.predict(tagged_table.features)
     assert True  # This asserts that the predict method succeeds
 
 
@@ -23,7 +23,7 @@ def test_gradient_boosting_predict_not_fitted() -> None:
     tagged_table = TaggedTable(table, "T")
     gradient_boosting = GradientBoosting()
     with pytest.raises(PredictionError):
-        gradient_boosting.predict(tagged_table.feature_vectors)
+        gradient_boosting.predict(tagged_table.features)
 
 
 def test_gradient_boosting_predict_invalid() -> None:
@@ -38,4 +38,4 @@ def test_gradient_boosting_predict_invalid() -> None:
     gradient_boosting = GradientBoosting()
     gradient_boosting.fit(tagged_table)
     with pytest.raises(PredictionError):
-        gradient_boosting.predict(invalid_tagged_table.feature_vectors)
+        gradient_boosting.predict(invalid_tagged_table.features)
diff --git a/tests/safeds/ml/classification/_k_nearest_neighbors/test_predict.py b/tests/safeds/ml/classification/_k_nearest_neighbors/test_predict.py
@@ -10,7 +10,7 @@ def test_k_nearest_neighbors_predict() -> None:
     tagged_table = TaggedTable(table, "T")
     k_nearest_neighbors = KNearestNeighborsClassifier(2)
     k_nearest_neighbors.fit(tagged_table)
-    k_nearest_neighbors.predict(tagged_table.feature_vectors)
+    k_nearest_neighbors.predict(tagged_table.features)
     assert True  # This asserts that the predict method succeeds
 
 
@@ -19,7 +19,7 @@ def test_k_nearest_neighbors_predict_not_fitted() -> None:
     tagged_table = TaggedTable(table, "T")
     k_nearest_neighbors = KNearestNeighborsClassifier(2)
     with pytest.raises(PredictionError):
-        k_nearest_neighbors.predict(tagged_table.feature_vectors)
+        k_nearest_neighbors.predict(tagged_table.features)
 
 
 def test_k_nearest_neighbors_predict_invalid() -> None:
@@ -32,4 +32,4 @@ def test_k_nearest_neighbors_predict_invalid() -> None:
     k_nearest_neighbors = KNearestNeighborsClassifier(2)
     k_nearest_neighbors.fit(tagged_table)
     with pytest.raises(PredictionError):
-        k_nearest_neighbors.predict(invalid_tagged_table.feature_vectors)
+        k_nearest_neighbors.predict(invalid_tagged_table.features)
diff --git a/tests/safeds/ml/classification/_logistic_regression/test_predict.py b/tests/safeds/ml/classification/_logistic_regression/test_predict.py
@@ -10,7 +10,7 @@ def test_logistic_regression_predict() -> None:
     tagged_table = TaggedTable(table, "T")
     log_regression = LogisticRegression()
     log_regression.fit(tagged_table)
-    log_regression.predict(tagged_table.feature_vectors)
+    log_regression.predict(tagged_table.features)
     assert True  # This asserts that the predict method succeeds
 
 
@@ -19,7 +19,7 @@ def test_logistic_regression_predict_not_fitted() -> None:
     tagged_table = TaggedTable(table, "T")
     log_regression = LogisticRegression()
     with pytest.raises(PredictionError):
-        log_regression.predict(tagged_table.feature_vectors)
+        log_regression.predict(tagged_table.features)
 
 
 def test_logistic_regression_predict_invalid() -> None:
@@ -32,4 +32,4 @@ def test_logistic_regression_predict_invalid() -> None:
     log_regression = LogisticRegression()
     log_regression.fit(tagged_table)
     with pytest.raises(PredictionError):
-        log_regression.predict(invalid_tagged_table.feature_vectors)
+        log_regression.predict(invalid_tagged_table.features)
diff --git a/tests/safeds/ml/classification/_random_forest/test_predict.py b/tests/safeds/ml/classification/_random_forest/test_predict.py
@@ -10,7 +10,7 @@ def test_random_forest_predict() -> None:
     tagged_table = TaggedTable(table, "T")
     random_forest = RandomForestClassifier()
     random_forest.fit(tagged_table)
-    random_forest.predict(tagged_table.feature_vectors)
+    random_forest.predict(tagged_table.features)
     assert True  # This asserts that the predict method succeeds
 
 
@@ -19,7 +19,7 @@ def test_random_forest_predict_not_fitted() -> None:
     tagged_table = TaggedTable(table, "T")
     random_forest = RandomForestClassifier()
     with pytest.raises(PredictionError):
-        random_forest.predict(tagged_table.feature_vectors)
+        random_forest.predict(tagged_table.features)
 
 
 def test_random_forest_predict_invalid() -> None:
@@ -32,4 +32,4 @@ def test_random_forest_predict_invalid() -> None:
     random_forest = RandomForestClassifier()
     random_forest.fit(tagged_table)
     with pytest.raises(PredictionError):
-        random_forest.predict(invalid_tagged_table.feature_vectors)
+        random_forest.predict(invalid_tagged_table.features)