CNN architecture review in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to add a convolutional layer with 32 filters and a 3x3 kernel.

Computer Vision

model.add(Conv2D([1], kernel_size=(3, 3), activation='relu', input_shape=(28, 28, 1)))

Drag options to blanks, or click blank then click option'

A32

B64

C16

D128

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing too few or too many filters like 16 or 128 which may underfit or overfit early.

Confusing kernel size with number of filters.

The first convolutional layer commonly uses 32 filters to extract features from the input images.

2fill in blank

medium

Complete the code to add a max pooling layer with pool size 2x2.

Computer Vision

model.add(MaxPooling2D(pool_size=[1]))

Drag options to blanks, or click blank then click option'

A(3, 3)

B(4, 4)

C(2, 2)

D(1, 1)

Attempts:

3 left

💡 Hint

Common Mistakes

Using (1, 1) which does not reduce size.

Using too large pool sizes like (4, 4) which may lose too much information.

Max pooling with a 2x2 window reduces the spatial size by half, helping reduce computation.

3fill in blank

hard

Fix the error in the code to flatten the output before the dense layer.

Computer Vision

model.add([1]())

Drag options to blanks, or click blank then click option'

AFlatten

BDense

CConv2D

DMaxPooling2D

Attempts:

3 left

💡 Hint

Common Mistakes

Using Dense directly without flattening causes shape errors.

Using Conv2D or MaxPooling2D here is incorrect.

Flatten layer converts 2D feature maps into 1D vectors for dense layers.

4fill in blank

hard

Fill both blanks to add a dropout layer with rate 0.5 and a dense output layer with 10 units.

Computer Vision

model.add(Dropout([1]))
model.add(Dense([2], activation='softmax'))

Drag options to blanks, or click blank then click option'

A0.5

B10

C5

D0.25

Attempts:

3 left

💡 Hint

Common Mistakes

Using dropout rates too low or too high.

Setting output units incorrectly for classification.

Dropout with 0.5 rate helps prevent overfitting. The output layer has 10 units for 10 classes.

5fill in blank

hard

Fill all three blanks to compile the model with Adam optimizer, categorical crossentropy loss, and accuracy metric.

Computer Vision

model.compile(optimizer='[1]', loss='[2]', metrics=['[3]'])

Drag options to blanks, or click blank then click option'

Aadam

Bcategorical_crossentropy

Caccuracy

Dsgd

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong loss function like 'mse' for classification.

Choosing incorrect optimizer or metric.

Adam optimizer is popular for CNNs. Categorical crossentropy is used for multi-class classification. Accuracy measures performance.