CNN architecture for image classification in PyTorch - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the main purpose of a Convolutional Neural Network (CNN) in image classification?

A CNN automatically learns to detect important features like edges, shapes, and textures from images to classify them into categories.

beginner

What does a convolutional layer do in a CNN?

It applies small filters to the input image to create feature maps that highlight important patterns like edges or textures.

beginner

Why do CNNs use pooling layers?

Pooling layers reduce the size of feature maps, making the model faster and helping it focus on the most important features.

beginner

What role does the fully connected layer play in a CNN for image classification?

It takes the extracted features and decides which class the image belongs to by combining all the information.

beginner

How is accuracy calculated during CNN training for image classification?

Accuracy is the percentage of images the CNN correctly classifies out of all images tested.

What is the first layer usually used in a CNN for image classification?

A Convolutional layer

B Pooling layer

C Fully connected layer

D Dropout layer

The convolutional layer is the first layer that extracts features from the input image.

Which layer reduces the spatial size of the feature maps?

A Convolutional layer

B Batch normalization layer

C Fully connected layer

D Pooling layer

Pooling layers reduce the size of feature maps to lower computation and focus on important features.

What does the output layer of a CNN for classification usually use?

A Softmax activation

B Sigmoid activation

C ReLU activation

D Tanh activation

Softmax activation converts outputs into probabilities for each class.

Which metric tells how many images were correctly classified?

A Loss

B Accuracy

C Precision

D Recall

Accuracy measures the percentage of correct predictions.

What is the main advantage of using convolutional layers over fully connected layers for images?

A They remove noise from images

B They increase the image size

C They reduce the number of parameters by sharing weights

D They convert images to text

Convolutional layers share weights across space, reducing parameters and improving efficiency.

Explain the main components of a CNN architecture used for image classification and their roles.

Think about how the network processes images step-by-step.

Describe how accuracy is calculated during CNN training and why it is important.

Consider what accuracy tells you about the model's performance.