import unittest
import importlib
from sklearn.model_selection import GridSearchCV
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score

def _dynamic_test(test_case, condition, success_msg, failure_msg):
    if condition:
        test_case._testMethodName = success_msg
        test_case.assertTrue(True, success_msg)
    else:
        test_case._testMethodName = failure_msg
        test_case.fail(failure_msg)


class TestGridSearch(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        cls.user = importlib.import_module("user_code")

    def test_grid_search_created(self):
        """Check if GridSearchCV object exists and is properly initialized."""
        grid_search = getattr(self.user, "grid_search", None)
        condition = isinstance(grid_search, GridSearchCV)
        _dynamic_test(
            self,
            condition,
            "GridSearchCV object successfully created.",
            "GridSearchCV object missing or incorrect type."
        )

    def test_param_grid_structure(self):
        """Check if parameter grid has the required keys."""
        grid_search = getattr(self.user, "grid_search", None)
        if grid_search:
            params = grid_search.param_grid
            expected_keys = {"n_estimators", "max_depth", "min_samples_split"}
            condition = expected_keys.issubset(params.keys())
        else:
            condition = False
        _dynamic_test(
            self,
            condition,
            "Parameter grid contains all required hyperparameters.",
            "Parameter grid missing one or more required keys."
        )

    def test_best_model_exists(self):
        """Check if best_model variable exists and is a RandomForestClassifier."""
        best_model = getattr(self.user, "best_model", None)
        condition = isinstance(best_model, RandomForestClassifier)
        _dynamic_test(
            self,
            condition,
            "best_model variable successfully defined as RandomForestClassifier.",
            "best_model missing or not a RandomForestClassifier."
        )

    def test_model_accuracy(self):
        """Check that test accuracy variable exists and is within reasonable range."""
        acc = getattr(self.user, "acc", None)
        condition = isinstance(acc, float) and 0.5 < acc <= 1.0
        _dynamic_test(
            self,
            condition,
            "Model accuracy variable defined and within valid range.",
            "Model accuracy missing or outside expected range (0.5–1.0)."
        )


if __name__ == "__main__":
    unittest.main(argv=["first-arg-is-ignored"], exit=False)


test_code.py

A comprehensive intermediate course guiding learners through the essentials of hyperparameter tuning in machine learning, from foundational theory to advanced automated techniques, with hands-on scikit-learn examples.

Build foundational understanding of hyperparameters, their distinction from parameters, and their critical role in model performance, generalization, and the bias-variance tradeoff.

Explore manual hyperparameter tuning, intuition-driven adjustments, and systematic search-based methods including grid and random search using scikit-learn.

Delve into advanced hyperparameter tuning strategies, including Bayesian optimization and best practices for reproducibility.

Challenge: Grid Search

Løsning

Challenge: Grid Search

Løsning