## Breast Cancer Dataset Overview

The **breast_cancer** dataset from scikit-learn is a widely used binary classification dataset for predicting whether a tumor is malignant or benign based on various features computed from digitized images of fine needle aspirate (FNA) of breast masses.

<div class="custom-table-wrapper">
  <link rel="stylesheet" href="https://staging-content-media-cdn.codefinity.com/courses/c1087363-9025-4c37-8d5e-983a32a007e9/tables-section/scrollable-framed-table.css" />
  <link rel="stylesheet" href="https://codefinity-content-media.s3.eu-west-1.amazonaws.com/css_custom_styles/TextFormatting.css" />
  <table>
    <tbody>
      <tr>
        <th><strong>Column Name</strong></th>
        <th><strong>Description</strong></th>
      </tr>
      <tr>
        <td><code>mean radius</code></td>
        <td>Mean of distances from center to points on the perimeter</td>
      </tr>
      <tr>
        <td><code>mean texture</code></td>
        <td>Standard deviation of gray-scale values</td>
      </tr>
      <tr>
        <td><code>mean perimeter</code></td>
        <td>Mean size of the perimeter</td>
      </tr>
      <tr>
        <td><code>mean area</code></td>
        <td>Mean area of the cell</td>
      </tr>
      <tr>
        <td><code>mean smoothness</code></td>
        <td>Mean of local variation in radius lengths</td>
      </tr>
      <tr>
        <td><code>mean compactness</code></td>
        <td>Mean of perimeter² / area - 1.0</td>
      </tr>
      <tr>
        <td><code>mean concavity</code></td>
        <td>Mean severity of concave portions of the contour</td>
      </tr>
      <tr>
        <td><code>mean concave points</code></td>
        <td>Mean number of concave portions of the contour</td>
      </tr>
      <tr>
        <td><code>mean symmetry</code></td>
        <td>Mean symmetry of the cell</td>
      </tr>
      <tr>
        <td><code>mean fractal dimension</code></td>
        <td>Mean "coastline approximation" - 1</td>
      </tr>
      <tr>
        <td><code>radius error</code></td>
        <td>Standard error of radius</td>
      </tr>
      <tr>
        <td><code>texture error</code></td>
        <td>Standard error of texture</td>
      </tr>
      <tr>
        <td><code>perimeter error</code></td>
        <td>Standard error of perimeter</td>
      </tr>
      <tr>
        <td><code>area error</code></td>
        <td>Standard error of area</td>
      </tr>
      <tr>
        <td><code>smoothness error</code></td>
        <td>Standard error of smoothness</td>
      </tr>
      <tr>
        <td><code>compactness error</code></td>
        <td>Standard error of compactness</td>
      </tr>
      <tr>
        <td><code>concavity error</code></td>
        <td>Standard error of concavity</td>
      </tr>
      <tr>
        <td><code>concave points error</code></td>
        <td>Standard error of concave points</td>
      </tr>
      <tr>
        <td><code>symmetry error</code></td>
        <td>Standard error of symmetry</td>
      </tr>
      <tr>
        <td><code>fractal dimension error</code></td>
        <td>Standard error of fractal dimension</td>
      </tr>
      <tr>
        <td><code>worst radius</code></td>
        <td>"Worst" or largest mean value for radius</td>
      </tr>
      <tr>
        <td><code>worst texture</code></td>
        <td>"Worst" or largest mean value for texture</td>
      </tr>
      <tr>
        <td><code>worst perimeter</code></td>
        <td>"Worst" or largest mean value for perimeter</td>
      </tr>
      <tr>
        <td><code>worst area</code></td>
        <td>"Worst" or largest mean value for area</td>
      </tr>
      <tr>
        <td><code>worst smoothness</code></td>
        <td>"Worst" or largest mean value for smoothness</td>
      </tr>
      <tr>
        <td><code>worst compactness</code></td>
        <td>"Worst" or largest mean value for compactness</td>
      </tr>
      <tr>
        <td><code>worst concavity</code></td>
        <td>"Worst" or largest mean value for concavity</td>
      </tr>
      <tr>
        <td><code>worst concave points</code></td>
        <td>"Worst" or largest mean value for concave points</td>
      </tr>
      <tr>
        <td><code>worst symmetry</code></td>
        <td>"Worst" or largest mean value for symmetry</td>
      </tr>
      <tr>
        <td><code>worst fractal dimension</code></td>
        <td>"Worst" or largest mean value for fractal dimension</td>
      </tr>
    </tbody>
  </table>
</div>

### Data Overview

- Number of samples: 569;
- Number of features: 30;
- Target variable: `target` (0 = malignant, 1 = benign);
- Task: Predict whether a tumor is malignant or benign based on the features above.


import unittest
import importlib
import numpy as np
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import (
    accuracy_score, precision_score, recall_score, f1_score,
    roc_auc_score, confusion_matrix
)

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

        data = load_breast_cancer()
        X_train, X_test, y_train, y_test = train_test_split(
            data.data, data.target, test_size=0.3, random_state=42
        )

        model = LogisticRegression(max_iter=1000)
        model.fit(X_train, y_train)
        y_pred = model.predict(X_test)
        y_prob = model.predict_proba(X_test)[:, 1]

        cls.true_metrics = {
            "accuracy": accuracy_score(y_test, y_pred),
            "precision": precision_score(y_test, y_pred),
            "recall": recall_score(y_test, y_pred),
            "f1": f1_score(y_test, y_pred),
            "roc_auc": roc_auc_score(y_test, y_prob),
            "cm": confusion_matrix(y_test, y_pred),
            "mean_cv": cross_val_score(model, data.data, data.target, cv=5).mean(),
        }

    def test_accuracy(self):
        cond = abs(self.user.accuracy - self.true_metrics["accuracy"]) < 1e-6
        _dynamic_test(self, cond, "Accuracy computed correctly.", "Accuracy is incorrect.")

    def test_precision(self):
        cond = abs(self.user.precision - self.true_metrics["precision"]) < 1e-6
        _dynamic_test(self, cond, "Precision computed correctly.", "Precision is incorrect.")

    def test_recall(self):
        cond = abs(self.user.recall - self.true_metrics["recall"]) < 1e-6
        _dynamic_test(self, cond, "Recall computed correctly.", "Recall is incorrect.")

    def test_f1(self):
        cond = abs(self.user.f1 - self.true_metrics["f1"]) < 1e-6
        _dynamic_test(self, cond, "F1 score computed correctly.", "F1 score is incorrect.")

    def test_roc_auc(self):
        cond = abs(self.user.roc_auc - self.true_metrics["roc_auc"]) < 1e-6
        _dynamic_test(self, cond, "ROC–AUC computed correctly.", "ROC–AUC is incorrect.")

    def test_confusion_matrix(self):
        cond = np.array_equal(self.user.cm, self.true_metrics["cm"])
        _dynamic_test(self, cond, "Confusion matrix computed correctly.", "Confusion matrix is incorrect.")

    def test_cross_validation(self):
        cond = abs(self.user.mean_cv - self.true_metrics["mean_cv"]) < 1e-6
        _dynamic_test(self, cond, "Cross-validation mean accuracy computed correctly.", "Cross-validation mean accuracy is incorrect.")

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

test_main.py

A comprehensive course for intermediate learners to master the key evaluation metrics used in machine learning, covering both supervised and unsupervised tasks. The course progresses from foundational concepts to practical Python implementations, interpretation of results, and a comparative overview of metric types and their trade-offs.

Explore the fundamental metrics used to evaluate classification models, including their definitions, formulas, Python implementations, and interpretation. Includes advanced metrics and model validation techniques relevant to classification.

Delve into the core metrics for evaluating regression models, including their mathematical foundations, Python implementations, and result interpretation. Includes advanced validation techniques relevant to regression.

Examine the key metrics for evaluating unsupervised learning models, including clustering, dimensionality reduction, and anomaly detection. Learn their mathematical foundations, Python implementations, and interpretation.

Challenge: Classification Metrics

Breast Cancer Dataset Overview

Data Overview

Lösning