Softmax output layer in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to add a softmax output layer with 3 classes.

TensorFlow

model.add(tf.keras.layers.Dense(3, activation=[1]))

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

A"relu"

B"tanh"

C"softmax"

D"sigmoid"

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'relu' or 'sigmoid' in the output layer for multi-class classification.

Forgetting to specify activation function.

The softmax activation function is used in the output layer for multi-class classification to produce probabilities for each class.

2fill in blank

medium

Complete the code to compile the model with the correct loss function for softmax output.

TensorFlow

model.compile(optimizer='adam', loss=[1], metrics=['accuracy'])

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

A"hinge"

B"binary_crossentropy"

C"mean_squared_error"

D"categorical_crossentropy"

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'binary_crossentropy' for multi-class problems.

Using loss functions meant for regression.

For multi-class classification with softmax output, 'categorical_crossentropy' is the appropriate loss function.

3fill in blank

hard

Fix the error in the code to correctly define a softmax output layer.

TensorFlow

output = tf.keras.layers.Dense(5, activation=[1])(previous_layer)

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

A"softmax"

Bsoftmax

C"Softmax"

D"relu"

Attempts:

3 left

💡 Hint

Common Mistakes

Passing activation without quotes.

Using uppercase letters in activation string.

Activation functions must be passed as lowercase strings like 'softmax'. Passing without quotes or with wrong case causes errors.

4fill in blank

hard

Fill both blanks to create a softmax output layer and compile the model correctly.

TensorFlow

model.add(tf.keras.layers.Dense([1], activation=[2]))
model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])

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

A10

B"softmax"

C5

D"sigmoid"

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong number of units for output layer.

Using 'sigmoid' instead of 'softmax' for multi-class output.

The output layer should have units equal to the number of classes (10 here) and use 'softmax' activation for multi-class classification.

5fill in blank

hard

Fill all three blanks to define a softmax output layer, compile the model, and fit it with data.

TensorFlow

model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(64, activation='relu'))
model.add(tf.keras.layers.Dense([1], activation=[2]))
model.compile(optimizer=[3], loss='categorical_crossentropy', metrics=['accuracy'])
model.fit(x_train, y_train, epochs=5)

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

A3

B"softmax"

C'adam'

D'sgd'

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong number of output units.

Using 'sigmoid' activation instead of 'softmax'.

Using an optimizer other than 'adam' without reason.

The output layer units match the number of classes (3), activation is 'softmax', and optimizer is 'adam' for good training performance.