feat: Improve Error Handling of classifiers and regressors (#355)

Closes #153 .

### Summary of Changes

The Error Handling of classifiers and regressors was not ideal. Now
every classifier and regressor validates the data, especially when given
non-numerical or missing values in the table.

<!-- Please provide a summary of changes in this pull request, ensuring
all changes are explained. -->

---------

Co-authored-by: Alexander Gréus <[email protected]>
Co-authored-by: megalinter-bot <[email protected]>
Co-authored-by: Severin Paul Höfer <[email protected]>
Co-authored-by: Alexander <[email protected]>
Co-authored-by: Junior Atemebang <[email protected]>

src/safeds/exceptions/__init__.py

@@ -5,6 +5,7 @@
     ColumnSizeError,
     DuplicateColumnNameError,
     IndexOutOfBoundsError,
+    MissingValuesColumnError,
     NonNumericColumnError,
     SchemaMismatchError,
     TransformerNotFittedError,
@@ -14,6 +15,7 @@
 )
 from safeds.exceptions._ml import (
     DatasetContainsTargetError,
+    DatasetMissesDataError,
     DatasetMissesFeaturesError,
     LearningError,
     ModelNotFittedError,
@@ -33,11 +35,13 @@
     "UnknownColumnNameError",
     "ValueNotPresentWhenFittedError",
     "WrongFileExtensionError",
+    "MissingValuesColumnError",
     # ML exceptions
     "DatasetContainsTargetError",
     "DatasetMissesFeaturesError",
     "LearningError",
     "ModelNotFittedError",
     "PredictionError",
     "UntaggedTableError",
+    "DatasetMissesDataError",
 ]

src/safeds/exceptions/_data.py

@@ -23,8 +23,27 @@ def __init__(self, column_names: list[str]):
 class NonNumericColumnError(Exception):
     """Exception raised for trying to do numerical operations on a non-numerical column."""
 
-    def __init__(self, column_info: str) -> None:
-        super().__init__(f"Tried to do a numerical operation on one or multiple non numerical Columns: \n{column_info}")
+    def __init__(self, column_info: str, help_msg: str | None = None) -> None:
+        line_break = "\n"
+        super().__init__(
+            (
+                "Tried to do a numerical operation on one or multiple non-numerical columns:"
+                f" \n{column_info}{line_break + help_msg if help_msg is not None else ''}"
+            ),
+        )
+
+
+class MissingValuesColumnError(Exception):
+    """Exception raised for trying to do operations on columns containing missing values."""
+
+    def __init__(self, column_info: str, help_msg: str | None = None) -> None:
+        line_break = "\n"
+        super().__init__(
+            (
+                "Tried to do an operation on one or multiple columns containing missing values:"
+                f" \n{column_info}{line_break + help_msg if help_msg is not None else ''}"
+            ),
+        )
 
 
 class DuplicateColumnNameError(Exception):

src/safeds/exceptions/_ml.py

@@ -26,6 +26,13 @@ def __init__(self, missing_feature_names: list[str]):
         super().__init__(f"Dataset misses the feature columns '{missing_feature_names}'.")
 
 
+class DatasetMissesDataError(ValueError):
+    """Raised when a dataset contains no rows."""
+
+    def __init__(self) -> None:
+        super().__init__("Dataset contains no rows")
+
+
 class LearningError(Exception):
     """
     Raised when an error occurred while training a model.

src/safeds/ml/classical/_util_sklearn.py

@@ -4,9 +4,12 @@
 from safeds.data.tabular.containers import Table, TaggedTable
 from safeds.exceptions import (
     DatasetContainsTargetError,
+    DatasetMissesDataError,
     DatasetMissesFeaturesError,
     LearningError,
+    MissingValuesColumnError,
     ModelNotFittedError,
+    NonNumericColumnError,
     PredictionError,
     UntaggedTableError,
 )
@@ -30,9 +33,44 @@ def fit(model: Any, tagged_table: TaggedTable) -> None:
         If the tagged table contains invalid values or if the training failed.
     UntaggedTableError
         If the table is untagged.
+    NonNumericColumnError
+        If the training data contains non-numerical values.
+    MissingValuesColumnError
+        If the training data contains missing values.
+    DatasetMissesDataError
+        If the training data contains no rows.
     """
     if not isinstance(tagged_table, TaggedTable) and isinstance(tagged_table, Table):
         raise UntaggedTableError
+
+    if tagged_table.number_of_rows == 0:
+        raise DatasetMissesDataError
+
+    non_numerical_column_names = set(tagged_table.features.column_names) - set(
+        tagged_table.features.remove_columns_with_non_numerical_values().column_names,
+    )
+    if len(non_numerical_column_names) != 0:
+        raise NonNumericColumnError(
+            str(non_numerical_column_names),
+            (
+                "You can use the LabelEncoder or OneHotEncoder to transform your non-numerical data to numerical"
+                " data.\nThe OneHotEncoder should be used if you work with nominal data. If your data contains too many"
+                " different values\nor is ordinal, you should use the LabelEncoder."
+            ),
+        )
+
+    null_containing_column_names = set(tagged_table.features.column_names) - set(
+        tagged_table.features.remove_columns_with_missing_values().column_names,
+    )
+    if len(null_containing_column_names) != 0:
+        raise MissingValuesColumnError(
+            str(null_containing_column_names),
+            (
+                "You can use the Imputer to replace the missing values based on different strategies.\nIf you want to"
+                " remove the missing values entirely you can use the method `Table.remove_rows_with_missing_values`."
+            ),
+        )
+
     try:
         model.fit(
             tagged_table.features._data,
@@ -73,6 +111,12 @@ def predict(model: Any, dataset: Table, feature_names: list[str] | None, target_
         If the dataset misses feature columns.
     PredictionError
         If predicting with the given dataset failed.
+    NonNumericColumnError
+        If the dataset contains non-numerical values.
+    MissingValuesColumnError
+        If the dataset contains missing values.
+    DatasetMissesDataError
+        If the dataset contains no rows.
     """
     # Validation
     if model is None or target_name is None or feature_names is None:
@@ -83,6 +127,34 @@ def predict(model: Any, dataset: Table, feature_names: list[str] | None, target_
     if missing_feature_names:
         raise DatasetMissesFeaturesError(missing_feature_names)
 
+    if dataset.number_of_rows == 0:
+        raise DatasetMissesDataError
+
+    non_numerical_column_names = set(dataset.keep_only_columns(feature_names).column_names) - set(
+        dataset.keep_only_columns(feature_names).remove_columns_with_non_numerical_values().column_names,
+    )
+    if len(non_numerical_column_names) != 0:
+        raise NonNumericColumnError(
+            str(non_numerical_column_names),
+            (
+                "You can use the LabelEncoder or OneHotEncoder to transform your non-numerical data to numerical"
+                " data.\nThe OneHotEncoder should be used if you work with nominal data. If your data contains too many"
+                " different values\nor is ordinal, you should use the LabelEncoder.\n"
+            ),
+        )
+
+    null_containing_column_names = set(dataset.keep_only_columns(feature_names).column_names) - set(
+        dataset.keep_only_columns(feature_names).remove_columns_with_missing_values().column_names,
+    )
+    if len(null_containing_column_names) != 0:
+        raise MissingValuesColumnError(
+            str(null_containing_column_names),
+            (
+                "You can use the Imputer to replace the missing values based on different strategies.\nIf you want to"
+                " remove the missing values entirely you can use the method `Table.remove_rows_with_missing_values`."
+            ),
+        )
+
     dataset_df = dataset.keep_only_columns(feature_names)._data
     dataset_df.columns = feature_names
 

src/safeds/ml/classical/classification/_ada_boost.py

@@ -90,6 +90,14 @@ def fit(self, training_set: TaggedTable) -> AdaBoost:
         ------
         LearningError
             If the training data contains invalid values or if the training failed.
+        UntaggedTableError
+            If the table is untagged.
+        NonNumericColumnError
+            If the training data contains non-numerical values.
+        MissingValuesColumnError
+            If the training data contains missing values.
+        DatasetMissesDataError
+            If the training data contains no rows.
         """
         wrapped_classifier = self._get_sklearn_classifier()
         fit(wrapped_classifier, training_set)
@@ -129,6 +137,12 @@ def predict(self, dataset: Table) -> TaggedTable:
             If the dataset misses feature columns.
         PredictionError
             If predicting with the given dataset failed.
+        NonNumericColumnError
+            If the dataset contains non-numerical values.
+        MissingValuesColumnError
+            If the dataset contains missing values.
+        DatasetMissesDataError
+            If the dataset contains no rows.
         """
         return predict(self._wrapped_classifier, dataset, self._feature_names, self._target_name)
 

src/safeds/ml/classical/classification/_decision_tree.py

@@ -43,6 +43,14 @@ def fit(self, training_set: TaggedTable) -> DecisionTree:
         ------
         LearningError
             If the training data contains invalid values or if the training failed.
+        UntaggedTableError
+            If the table is untagged.
+        NonNumericColumnError
+            If the training data contains non-numerical values.
+        MissingValuesColumnError
+            If the training data contains missing values.
+        DatasetMissesDataError
+            If the training data contains no rows.
         """
         wrapped_classifier = self._get_sklearn_classifier()
         fit(wrapped_classifier, training_set)
@@ -78,6 +86,12 @@ def predict(self, dataset: Table) -> TaggedTable:
             If the dataset misses feature columns.
         PredictionError
             If predicting with the given dataset failed.
+        NonNumericColumnError
+            If the dataset contains non-numerical values.
+        MissingValuesColumnError
+            If the dataset contains missing values.
+        DatasetMissesDataError
+            If the dataset contains no rows.
         """
         return predict(self._wrapped_classifier, dataset, self._feature_names, self._target_name)
 

src/safeds/ml/classical/classification/_gradient_boosting.py

@@ -77,6 +77,14 @@ def fit(self, training_set: TaggedTable) -> GradientBoosting:
         ------
         LearningError
             If the training data contains invalid values or if the training failed.
+        UntaggedTableError
+            If the table is untagged.
+        NonNumericColumnError
+            If the training data contains non-numerical values.
+        MissingValuesColumnError
+            If the training data contains missing values.
+        DatasetMissesDataError
+            If the training data contains no rows.
         """
         wrapped_classifier = self._get_sklearn_classifier()
         fit(wrapped_classifier, training_set)
@@ -112,6 +120,12 @@ def predict(self, dataset: Table) -> TaggedTable:
             If the dataset misses feature columns.
         PredictionError
             If predicting with the given dataset failed.
+        NonNumericColumnError
+            If the dataset contains non-numerical values.
+        MissingValuesColumnError
+            If the dataset contains missing values.
+        DatasetMissesDataError
+            If the dataset contains no rows.
         """
         return predict(self._wrapped_classifier, dataset, self._feature_names, self._target_name)
 

src/safeds/ml/classical/classification/_k_nearest_neighbors.py

@@ -4,6 +4,7 @@
 
 from sklearn.neighbors import KNeighborsClassifier as sk_KNeighborsClassifier
 
+from safeds.exceptions import DatasetMissesDataError
 from safeds.ml.classical._util_sklearn import fit, predict
 
 from ._classifier import Classifier
@@ -69,7 +70,17 @@ def fit(self, training_set: TaggedTable) -> KNearestNeighbors:
             If `number_of_neighbors` is greater than the sample size.
         LearningError
             If the training data contains invalid values or if the training failed.
+        UntaggedTableError
+            If the table is untagged.
+        NonNumericColumnError
+            If the training data contains non-numerical values.
+        MissingValuesColumnError
+            If the training data contains missing values.
+        DatasetMissesDataError
+            If the training data contains no rows.
         """
+        if training_set.number_of_rows == 0:
+            raise DatasetMissesDataError
         if self._number_of_neighbors > training_set.number_of_rows:
             raise ValueError(
                 (
@@ -111,6 +122,12 @@ def predict(self, dataset: Table) -> TaggedTable:
             If the dataset misses feature columns.
         PredictionError
             If predicting with the given dataset failed.
+        NonNumericColumnError
+            If the dataset contains non-numerical values.
+        MissingValuesColumnError
+            If the dataset contains missing values.
+        DatasetMissesDataError
+            If the dataset contains no rows.
         """
         return predict(self._wrapped_classifier, dataset, self._feature_names, self._target_name)
 

src/safeds/ml/classical/classification/_logistic_regression.py

@@ -43,6 +43,14 @@ def fit(self, training_set: TaggedTable) -> LogisticRegression:
         ------
         LearningError
             If the training data contains invalid values or if the training failed.
+        UntaggedTableError
+            If the table is untagged.
+        NonNumericColumnError
+            If the training data contains non-numerical values.
+        MissingValuesColumnError
+            If the training data contains missing values.
+        DatasetMissesDataError
+            If the training data contains no rows.
         """
         wrapped_classifier = self._get_sklearn_classifier()
         fit(wrapped_classifier, training_set)
@@ -78,6 +86,12 @@ def predict(self, dataset: Table) -> TaggedTable:
             If the dataset misses feature columns.
         PredictionError
             If predicting with the given dataset failed.
+        NonNumericColumnError
+            If the dataset contains non-numerical values.
+        MissingValuesColumnError
+            If the dataset contains missing values.
+        DatasetMissesDataError
+            If the dataset contains no rows.
         """
         return predict(self._wrapped_classifier, dataset, self._feature_names, self._target_name)
 

src/safeds/ml/classical/classification/_random_forest.py

@@ -65,6 +65,14 @@ def fit(self, training_set: TaggedTable) -> RandomForest:
         ------
         LearningError
             If the training data contains invalid values or if the training failed.
+        UntaggedTableError
+            If the table is untagged.
+        NonNumericColumnError
+            If the training data contains non-numerical values.
+        MissingValuesColumnError
+            If the training data contains missing values.
+        DatasetMissesDataError
+            If the training data contains no rows.
         """
         wrapped_classifier = self._get_sklearn_classifier()
         fit(wrapped_classifier, training_set)
@@ -100,6 +108,12 @@ def predict(self, dataset: Table) -> TaggedTable:
             If the dataset misses feature columns.
         PredictionError
             If predicting with the given dataset failed.
+        NonNumericColumnError
+            If the dataset contains non-numerical values.
+        MissingValuesColumnError
+            If the dataset contains missing values.
+        DatasetMissesDataError
+            If the dataset contains no rows.
         """
         return predict(self._wrapped_classifier, dataset, self._feature_names, self._target_name)
 

src/safeds/ml/classical/classification/_support_vector_machine.py

@@ -119,6 +119,14 @@ def fit(self, training_set: TaggedTable) -> SupportVectorMachine:
         ------
         LearningError
             If the training data contains invalid values or if the training failed.
+        UntaggedTableError
+            If the table is untagged.
+        NonNumericColumnError
+            If the training data contains non-numerical values.
+        MissingValuesColumnError
+            If the training data contains missing values.
+        DatasetMissesDataError
+            If the training data contains no rows.
         """
         wrapped_classifier = self._get_sklearn_classifier()
         fit(wrapped_classifier, training_set)
@@ -154,6 +162,12 @@ def predict(self, dataset: Table) -> TaggedTable:
             If the dataset misses feature columns.
         PredictionError
             If predicting with the given dataset failed.
+        NonNumericColumnError
+            If the dataset contains non-numerical values.
+        MissingValuesColumnError
+            If the dataset contains missing values.
+        DatasetMissesDataError
+            If the dataset contains no rows.
         """
         return predict(self._wrapped_classifier, dataset, self._feature_names, self._target_name)