Convolutional Neural Networks Quiz

An input volume of size 32x32x3 is passed through a convolutional layer with 10 filters of size 5x5, a stride of 1, and 'VALID' (zero) padding. What are the dimensions of the output volume?

• 32x32x10

• 28x28x10

• 27x27x10

• 32x32x5

An input volume of size 64x64x3 is passed through a convolutional layer with 16 filters of size 3x3, a stride of 1, and 'SAME' padding. What are the dimensions of the output volume?

• 62x62x16

• 64x64x16

• 64x64x3

• 32x32x16

A feature map of size 56x56x128 is passed through a Max Pooling layer with a 2x2 window and a stride of 2. What are the dimensions of the output volume?

• 56x56x64

• 28x28x128

• 28x28x64

• 55x55x128

A convolutional layer uses a 7x7 filter, a stride of 2, and 3 units of padding on each side. If the input width is 64, what is the output width?

• 31

• 32

• 33

• 64

An input feature map has dimensions 14x14x256. It is passed to a Global Average Pooling (GAP) layer. What are the dimensions of the GAP layer's output?

• 1x1x256

• 14x14x1

• 1x1x1

• The operation is not possible.

A 4x4 feature map is fed into a Transposed Convolution layer with a 3x3 kernel, a stride of 2, and no padding. What are the spatial dimensions of the output?

• 5x5

• 8x8

• 9x9

• 10x10

A convolutional layer takes a 64x64x16 input volume and uses 32 filters, each of size 3x3. What is the total number of learnable parameters (weights AND biases) in this layer?

• 4,608

• 4,640

• 1,536

• 1,568

Consider a convolutional layer with 64 filters of size 3x3. The input to this layer has a depth of 32 channels. What is the total number of weight parameters in this layer (excluding biases)?

• 576

• 18,432

• 2,048

• 18,496

A feature map of size 7x7x512 is flattened and then passed to a fully connected (Dense) layer with 4096 neurons. How many weight parameters are there in this Dense layer?

• (7 * 7 * 512) + 4096

• (7 * 7 * 512) * 4096

• 512 * 4096

• 7 * 7 * 4096

A model has an input of 28x28x128. It is processed by a 1x1 convolution with 32 filters. How many parameters (weights and biases) does this 1x1 convolution layer have?

• 4,096

• 4,128

• 32

• 128 * 32