Model evaluation best practices in Computer Vision - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding Overfitting in Model Evaluation

Which of the following best describes the problem of overfitting during model evaluation?

A The model performs poorly on both training and test data.

B The model performs well on test data but poorly on training data.

C The model performs well on training data but poorly on new, unseen data.

D The model performs equally well on training and test data.

💡 Hint

Think about when a model memorizes training data but fails to generalize.

Overfitting means the model learns the training data too well, including noise, so it fails on new data.

❓ Metrics

intermediate

Choosing the Right Metric for Imbalanced Data

You have a binary classification model for detecting rare diseases in images. Which metric is most appropriate to evaluate the model?

A Accuracy

B F1 Score

C Precision

D Recall

💡 Hint

Consider a metric that balances false positives and false negatives.

F1 Score balances precision and recall, making it suitable for imbalanced classes.

❓ Predict Output

advanced

Output of Cross-Validation Accuracy Calculation

What is the output of the following Python code snippet?

Computer Vision

from sklearn.datasets import load_digits
from sklearn.model_selection import cross_val_score
from sklearn.svm import SVC

data = load_digits()
X, y = data.data, data.target
model = SVC(kernel='linear', random_state=42)
scores = cross_val_score(model, X, y, cv=5)
print(round(scores.mean(), 2))

A 0.95

B 0.85

C 0.75

D 0.65

💡 Hint

The digits dataset is relatively easy to classify with a linear SVM.

The linear SVM on the digits dataset typically achieves around 95% accuracy with 5-fold cross-validation.

🔧 Debug

advanced

Identifying the Bug in Model Evaluation Code

What error will this code raise when evaluating a classification model?

Computer Vision

from sklearn.metrics import accuracy_score

y_true = [0, 1, 2, 2, 1]
y_pred = [0, 2, 1, 2, 0]

score = accuracy_score(y_true, y_pred, average='macro')
print(score)

A TypeError: accuracy_score() got an unexpected keyword argument 'average'

B ValueError: Found input variables with inconsistent numbers of samples

C 0.4

D NameError: name 'average' is not defined

💡 Hint

Check the parameters accepted by accuracy_score.

accuracy_score does not accept an 'average' parameter; it causes a TypeError.

❓ Model Choice

expert

Best Model Choice for Multi-Class Image Classification with Limited Data

You want to build a model to classify images into 10 categories. You have only 500 labeled images. Which approach is best for evaluation and model choice?

A Train a deep CNN from scratch and evaluate on a single 80/20 train-test split.

B Use k-means clustering to label images and evaluate with silhouette score.

C Train a simple logistic regression on raw pixels and evaluate using accuracy on training data.

D Use transfer learning with a pre-trained CNN and evaluate using stratified 5-fold cross-validation.

💡 Hint

Consider data size and evaluation robustness.

Transfer learning leverages pre-trained features, and stratified cross-validation ensures balanced evaluation on limited data.