Classification Accuracy Metric

In machine learning classification tasks, evaluating model performance is essential to understanding how well our models generalize to unseen data. Among the various evaluation metrics available, accuracy stands as one of the most intuitive and widely-used measures for classification problems.

Accuracy represents the proportion of predictions that match the true labels. It answers a simple yet powerful question: "Out of all the predictions made, how many were correct?"

Given two arrays of equal length:

• y_actual: The ground truth labels (what the correct classifications actually are)
• y_predicted: The model's predicted labels (what the classifier predicted)

The accuracy is calculated using the formula:

$$\text{Accuracy} = \frac{\text{Number of Correct Predictions}}{\text{Total Number of Predictions}} = \frac{\sum_{i=1}^{n} \mathbb{1}(y_{actual_i} = y_{predicted_i})}{n}$$

Where 𝕝(condition) is the indicator function that returns 1 if the condition is true and 0 otherwise, and n is the total number of samples.

Understanding Accuracy Values:

• An accuracy of 1.0 (100%) means every prediction matched the actual label — a perfect classifier
• An accuracy of 0.0 (0%) means no predictions matched — the worst possible performance
• An accuracy of 0.5 (50%) on a binary classification task corresponds to random guessing

Your Task: Write a Python function that computes the classification accuracy given two numpy arrays containing the actual and predicted labels. The function should:

• Compare each corresponding pair of actual and predicted labels
• Count the number of matching (correct) predictions
• Return the accuracy as a floating-point value between 0 and 1