Data Science and Machine Learning (Theory and Projects) A to Z - Deep Neural Network Architecture Activity

The video tutorial explains the process of building a convolutional neural network model. It starts with defining the initial image dimensions and setting up the model. Each convolutional layer is followed by a max pooling layer, which reduces the spatial dimensions by half while doubling the number of channels. This process continues until the spatial dimensions are less than ten. At this point, a fully connected layer with 1000 units is added, followed by a softmax layer. The tutorial concludes with instructions on calculating the total number of parameters in the model.

What is the initial size of the RGB images used in the convolutional neural network model?

• 112x112x3

• 224x224x3

• 56x56x3

• 128x128x3

What happens to the spatial dimensions after each max pooling layer in the model?

• They shrink by half.

• They increase by a factor of three.

• They remain the same.

• They double in size.

How do the number of channels change after each convolutional and pooling layer block?

• They double.

• They triple.

• They remain constant.

• They decrease by half.

At what point should you stop adding convolutional and pooling layers to the model?

• When the spatial dimensions are smaller than ten.

• When the number of channels exceeds 1000.

• When the spatial dimensions are exactly 20.

• When the model has 10 layers.

What is the purpose of the fully connected layer in the model?

• To reduce the number of channels.

• To classify the input into categories.

• To increase the spatial dimensions.

• To perform data augmentation.