feat: plot decision tree (#876)

Closes #856 

### Summary of Changes

- Added plot method for DecisionTreeClassifier and
DecisionTreeRegressor.
- Added tests for both

---------

Co-authored-by: megalinter-bot <[email protected]>
Co-authored-by: Saman Hushi <[email protected]>
Co-authored-by: peplaul0 <[email protected]>
Co-authored-by: Lars Reimann <[email protected]>

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

@@ -3,6 +3,8 @@
 from typing import TYPE_CHECKING
 
 from safeds._utils import _structural_hash
+from safeds.data.image.containers import Image
+from safeds.exceptions._ml import ModelNotFittedError
 from safeds.ml.classical._bases import _DecisionTreeBase
 
 from ._classifier import Classifier
@@ -71,3 +73,41 @@ def _get_sklearn_model(self) -> ClassifierMixin:
             max_depth=self._max_depth,
             min_samples_leaf=self._min_sample_count_in_leaves,
         )
+
+    # ------------------------------------------------------------------------------------------------------------------
+    # Plot
+    # ------------------------------------------------------------------------------------------------------------------
+
+    def plot(self) -> Image:
+        """
+        Get the image of the decision tree.
+
+        Returns
+        -------
+        plot:
+            The decision tree figure as an image.
+
+        Raises
+        ------
+        ModelNotFittedError:
+            If model is not fitted.
+        """
+        if not self.is_fitted:
+            raise ModelNotFittedError
+
+        from io import BytesIO
+
+        import matplotlib.pyplot as plt
+        from sklearn.tree import plot_tree
+
+        plot_tree(self._wrapped_model)
+
+        # save plot fig bytes in buffer
+        with BytesIO() as buffer:
+            plt.savefig(buffer)
+            image = buffer.getvalue()
+
+        # prevent forced plot from sklearn showing
+        plt.close()
+
+        return Image.from_bytes(image)

src/safeds/ml/classical/regression/_decision_tree_regressor.py

@@ -3,6 +3,8 @@
 from typing import TYPE_CHECKING
 
 from safeds._utils import _structural_hash
+from safeds.data.image.containers import Image
+from safeds.exceptions._ml import ModelNotFittedError
 from safeds.ml.classical._bases import _DecisionTreeBase
 
 from ._regressor import Regressor
@@ -71,3 +73,41 @@ def _get_sklearn_model(self) -> RegressorMixin:
             max_depth=self._max_depth,
             min_samples_leaf=self._min_sample_count_in_leaves,
         )
+
+    # ------------------------------------------------------------------------------------------------------------------
+    # Plot
+    # ------------------------------------------------------------------------------------------------------------------
+
+    def plot(self) -> Image:
+        """
+        Get the image of the decision tree.
+
+        Returns
+        -------
+        plot:
+            The decision tree figure as an image.
+
+        Raises
+        ------
+        ModelNotFittedError:
+            If model is not fitted.
+        """
+        if not self.is_fitted:
+            raise ModelNotFittedError
+
+        from io import BytesIO
+
+        import matplotlib.pyplot as plt
+        from sklearn.tree import plot_tree
+
+        plot_tree(self._wrapped_model)
+
+        # save plot fig bytes in buffer
+        with BytesIO() as buffer:
+            plt.savefig(buffer)
+            image = buffer.getvalue()
+
+        # prevent forced plot from sklearn showing
+        plt.close()
+
+        return Image.from_bytes(image)

tests/safeds/ml/classical/classification/test_decision_tree.py

@@ -1,8 +1,11 @@
 import pytest
 from safeds.data.labeled.containers import TabularDataset
 from safeds.data.tabular.containers import Table
-from safeds.exceptions import OutOfBoundsError
+from safeds.exceptions import ModelNotFittedError, OutOfBoundsError
 from safeds.ml.classical.classification import DecisionTreeClassifier
+from syrupy import SnapshotAssertion
+
+from tests.helpers import os_mac, skip_if_os
 
 
 @pytest.fixture()
@@ -41,3 +44,21 @@ def test_should_be_passed_to_sklearn(self, training_set: TabularDataset) -> None
     def test_should_raise_if_less_than_or_equal_to_0(self, min_sample_count_in_leaves: int) -> None:
         with pytest.raises(OutOfBoundsError):
             DecisionTreeClassifier(min_sample_count_in_leaves=min_sample_count_in_leaves)
+
+
+class TestPlot:
+    def test_should_raise_if_model_is_not_fitted(self) -> None:
+        model = DecisionTreeClassifier()
+        with pytest.raises(ModelNotFittedError):
+            model.plot()
+
+    def test_should_check_that_plot_image_is_same_as_snapshot(
+        self,
+        training_set: TabularDataset,
+        snapshot_png_image: SnapshotAssertion,
+    ) -> None:
+        skip_if_os([os_mac])
+
+        fitted_model = DecisionTreeClassifier().fit(training_set)
+        image = fitted_model.plot()
+        assert image == snapshot_png_image

tests/safeds/ml/classical/regression/test_decision_tree.py

@@ -1,8 +1,9 @@
 import pytest
 from safeds.data.labeled.containers import TabularDataset
 from safeds.data.tabular.containers import Table
-from safeds.exceptions import OutOfBoundsError
+from safeds.exceptions import ModelNotFittedError, OutOfBoundsError
 from safeds.ml.classical.regression import DecisionTreeRegressor
+from syrupy import SnapshotAssertion
 
 
 @pytest.fixture()
@@ -41,3 +42,19 @@ def test_should_be_passed_to_sklearn(self, training_set: TabularDataset) -> None
     def test_should_raise_if_less_than_or_equal_to_0(self, min_sample_count_in_leaves: int) -> None:
         with pytest.raises(OutOfBoundsError):
             DecisionTreeRegressor(min_sample_count_in_leaves=min_sample_count_in_leaves)
+
+
+class TestPlot:
+    def test_should_raise_if_model_is_not_fitted(self) -> None:
+        model = DecisionTreeRegressor()
+        with pytest.raises(ModelNotFittedError):
+            model.plot()
+
+    def test_should_check_that_plot_image_is_same_as_snapshot(
+        self,
+        training_set: TabularDataset,
+        snapshot_png_image: SnapshotAssertion,
+    ) -> None:
+        fitted_model = DecisionTreeRegressor().fit(training_set)
+        image = fitted_model.plot()
+        assert image == snapshot_png_image