RHOAIENG-26843, RHOAIENG-26066: tests: add scikit-learn smoke test (#1171)

* feat: Add scikit-learn smoke test

Adds a smoke test to verify basic scikit-learn functionality
in JupyterLab images.

This test trains a simple Logistic Regression model and checks
predictions to ensure the scikit-learn installation is working
correctly after package updates.

* RHOAIENG-27131: fix PEP 8 spacing issues in JupyterLab test

* RHOAIENG-26843: update sklearn smoke test with refined allure issue link and minor script improvements

* RHOAIENG-26843: add pytest fixture for JupyterLab datascience images and move sklearn smoke test to dedicated test file

* RHOAIENG-27131: fix PEP 8 formatting in datascience image fixture

* RHOAIENG-27141: add NumPy version logging to JupyterLab datascience test

* RHOAIENG-27142: add random seed to sklearn test for reproducibility

---------

Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>

Author: jiridanek

tests/containers/conftest.py

@@ -152,6 +152,16 @@ def jupyterlab_image(image: str) -> Image:
     return image_metadata
 
 
+@pytest.fixture(scope="function")
+def jupyterlab_datascience_image(jupyterlab_image: Image) -> Image:
+    if "-minimal-" in jupyterlab_image.labels["name"]:
+        pytest.skip(
+            f"Image {jupyterlab_image.name} is not datascience image because it has '-minimal-' in {jupyterlab_image.labels['name']=}'"
+        )
+
+    return jupyterlab_image
+
+
 @pytest.fixture(scope="function")
 def rstudio_image(image: str) -> Image:
     image_metadata = skip_if_not_workbench_image(image)

tests/containers/workbenches/jupyterlab/jupyterlab_datascience_test.py

@@ -0,0 +1,72 @@
+from __future__ import annotations
+
+import pathlib
+import tempfile
+
+import allure
+
+from tests.containers import conftest, docker_utils
+from tests.containers.workbenches.workbench_image_test import WorkbenchContainer
+
+
+class TestJupyterLabDatascienceImage:
+    """Tests for JupyterLab Workbench images in this repository that are not -minimal-."""
+
+    APP_ROOT_HOME = "/opt/app-root/src"
+
+    @allure.issue("RHOAIENG-26843")
+    @allure.description("Check that basic scikit-learn functionality is working.")
+    def test_sklearn_smoke(self, jupyterlab_datascience_image: conftest.Image) -> None:
+        container = WorkbenchContainer(image=jupyterlab_datascience_image.name, user=4321, group_add=[0])
+        # language=Python
+        test_script_content = """
+import sklearn
+from sklearn.linear_model import LogisticRegression
+import numpy as np
+
+# Set random seed for reproducibility
+np.random.seed(42)
+
+# Simple dataset
+X = np.array([[1], [2], [3], [4], [5]])
+y = np.array([0, 0, 1, 1, 1])
+
+# Train a model
+model = LogisticRegression(solver='liblinear', random_state=42)
+model.fit(X, y)
+
+# Make a prediction
+pred = model.predict([[3.5]])
+print(f"NumPy version: {np.__version__}")
+print(f"Scikit-learn version: {sklearn.__version__}")
+print(f"Prediction: {pred}")
+# We expect class 1 for input 3.5
+assert pred[0] == 1, "Prediction is not as expected"
+
+print("Scikit-learn smoke test completed successfully.")
+"""
+        test_script_name = "test_sklearn.py"
+        try:
+            container.start(wait_for_readiness=True)
+            with tempfile.TemporaryDirectory() as tmpdir:
+                tmpdir_path = pathlib.Path(tmpdir)
+                script_path = tmpdir_path / test_script_name
+                script_path.write_text(test_script_content)
+                docker_utils.container_cp(
+                    container.get_wrapped_container(),
+                    src=str(script_path),
+                    dst=self.APP_ROOT_HOME,
+                )
+
+            script_container_path = f"{self.APP_ROOT_HOME}/{test_script_name}"
+            exit_code, output = container.exec(["python", script_container_path])
+            output_str = output.decode()
+
+            print(f"Script output:\n{output_str}")
+
+            assert exit_code == 0, f"Script execution failed with exit code {exit_code}. Output:\n{output_str}"
+            assert "Scikit-learn smoke test completed successfully." in output_str
+            assert "Prediction: [1]" in output_str
+
+        finally:
+            docker_utils.NotebookContainer(container).stop(timeout=0)