Conditional generation in TensorFlow - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Model Choice

intermediate

Choosing the right model for conditional generation

You want to generate images conditioned on class labels using TensorFlow. Which model architecture is best suited for this task?

A A standard GAN without any label input

B A Variational Autoencoder (VAE) without conditioning

C A Conditional GAN (cGAN) that takes class labels as input to both generator and discriminator

D A simple feedforward neural network for classification

💡 Hint

Think about how the model can use label information to guide generation.

Conditional GANs (cGANs) explicitly use class labels as input to both generator and discriminator, enabling the generator to produce images conditioned on those labels. Standard GANs and VAEs without conditioning cannot control output classes. A feedforward network is for classification, not generation.

❓ Predict Output

intermediate

Output shape of conditional generator

Consider this TensorFlow code snippet for a conditional generator model:

import tensorflow as tf

noise = tf.random.normal([16, 100])
labels = tf.one_hot(tf.random.uniform([16], maxval=10, dtype=tf.int32), depth=10)
input_vector = tf.concat([noise, labels], axis=1)

model = tf.keras.Sequential([
    tf.keras.layers.Dense(128, activation='relu'),
    tf.keras.layers.Dense(28 * 28 * 1, activation='sigmoid'),
    tf.keras.layers.Reshape((28, 28, 1))
])

output = model(input_vector)
print(output.shape)

What will be printed?

A (16, 28, 28, 1)

B (100, 28, 28, 1)

C (16, 784)

D (16, 138)

💡 Hint

Check the batch size and the final reshape layer.

The input batch size is 16. The model outputs a tensor reshaped to (28, 28, 1) per example. So the output shape is (16, 28, 28, 1).

❓ Hyperparameter

advanced

Choosing the right loss function for conditional GAN training

When training a conditional GAN in TensorFlow, which loss function combination is most appropriate to encourage the generator to produce realistic images conditioned on labels?

A Huber loss for discriminator and Mean Absolute Error for generator

B Mean Squared Error (MSE) loss for both generator and discriminator

C Categorical Cross-Entropy loss only on the generator

D Binary Cross-Entropy loss for discriminator and generator with label smoothing

💡 Hint

Think about the typical GAN loss and how label smoothing helps.

Binary Cross-Entropy loss is standard for GANs to distinguish real vs fake. Label smoothing helps stabilize training by preventing the discriminator from becoming too confident. MSE or Huber losses are not typical for GAN adversarial losses.

❓ Metrics

advanced

Evaluating conditional generation quality

Which metric is best suited to quantitatively evaluate the quality and diversity of images generated by a conditional GAN trained on labeled image data?

A Inception Score (IS) computed on generated images

B Structural Similarity Index (SSIM) between generated and real images

C Accuracy of a classifier trained on generated images only

D Mean Squared Error (MSE) between generated and real images

💡 Hint

Consider a metric that measures both quality and diversity without needing paired images.

Inception Score evaluates both the quality and diversity of generated images using a pretrained classifier. MSE and SSIM require paired images and do not measure diversity. Classifier accuracy on generated images alone does not reflect generation quality.

🔧 Debug

expert

Debugging mode collapse in conditional GAN training

You trained a conditional GAN in TensorFlow, but the generator produces nearly identical images regardless of input labels. What is the most likely cause?

A The learning rate is too low causing slow convergence

B The generator input noise vector is not concatenated with the label embedding

C The discriminator uses label information but the generator does not

D The batch size is too large causing overfitting

💡 Hint

Think about how the generator receives conditioning information.

If the generator does not receive label information concatenated with noise, it cannot learn to produce different outputs conditioned on labels, causing mode collapse. The discriminator alone having labels is insufficient. Batch size and learning rate issues cause other problems but not this specific failure.

Practice

1. What is the main purpose of conditional generation in TensorFlow models?

easy

A. To generate outputs based on extra input information

B. To train models without any input data

C. To reduce the size of the model

D. To speed up the training process

Solution

1. Step 1: Understand conditional generation concept

   Conditional generation uses extra input information to guide the output generation.

2. Step 2: Identify the purpose in TensorFlow models

   This extra input helps models create outputs that depend on given conditions, like labels or attributes.

3. Final Answer:

   To generate outputs based on extra input information -> Option A

4. Quick Check:

   Conditional generation = guided output [OK]

Hint: Conditional means output depends on extra input [OK]

Common Mistakes:

• Thinking conditional generation removes input data
• Confusing conditional generation with model size reduction
• Assuming it only speeds up training

2. Which of the following is the correct way to combine noise and condition input in TensorFlow for conditional generation?

easy

A. tf.concat([noise, condition], axis=1)

B. tf.add(noise, condition)

C. tf.multiply(noise, condition)

D. tf.subtract(condition, noise)

Solution

1. Step 1: Understand input combination for conditional generation

   Noise and condition inputs are usually combined by concatenation along the feature axis.

2. Step 2: Identify correct TensorFlow function

   tf.concat joins tensors along a specified axis, which is the right way to combine noise and condition vectors.

3. Final Answer:

   tf.concat([noise, condition], axis=1) -> Option A

4. Quick Check:

   Combine inputs by concatenation [OK]

Hint: Use tf.concat to join inputs, not arithmetic ops [OK]

Common Mistakes:

• Using tf.add or tf.multiply which require same shapes
• Subtracting inputs which is not standard for combining
• Not specifying axis in concatenation

3. Given the code below, what will be the shape of the combined input tensor?

noise = tf.random.normal([32, 100])
condition = tf.random.normal([32, 10])
combined = tf.concat([noise, condition], axis=1)

medium

A. (32, 90)

B. (100, 10)

C. (32, 110)

D. (42, 110)

Solution

1. Step 1: Check shapes of noise and condition tensors

   Noise shape is (32, 100), condition shape is (32, 10).

2. Step 2: Understand concatenation along axis=1

   Concatenating along axis=1 adds the second dimension sizes: 100 + 10 = 110, batch size 32 stays same.

3. Final Answer:

   (32, 110) -> Option C

4. Quick Check:

   Concatenate dims: 100+10=110 [OK]

Hint: Concatenate adds sizes on axis=1, batch size unchanged [OK]

Common Mistakes:

• Adding batch sizes instead of feature sizes
• Mixing up axis parameter in concat
• Assuming shapes transpose automatically

4. Identify the error in this conditional generation code snippet:

noise = tf.random.normal([32, 100])
condition = tf.random.normal([10, 32])
combined = tf.concat([noise, condition], axis=1)

medium

A. Axis parameter should be 0

B. tf.concat cannot concatenate tensors

C. Noise tensor shape is invalid

D. Shapes of noise and condition do not match on axis 0

Solution

1. Step 1: Compare shapes of noise and condition tensors

   Noise shape is (32, 100), condition shape is (10, 32). Batch sizes differ (32 vs 10).

2. Step 2: Understand concat axis requirement

   Concatenation along axis=1 requires matching sizes on axis=0 (batch size). Here they differ, causing error.

3. Final Answer:

   Shapes of noise and condition do not match on axis 0 -> Option D

4. Quick Check:

   Batch sizes must match for concat [OK]

Hint: Check batch sizes match before concatenation [OK]

Common Mistakes:

• Assuming tf.concat works with any shapes
• Changing axis without fixing shape mismatch
• Ignoring batch size dimension

5. You want to build a conditional GAN in TensorFlow that generates images of specific digits (0-9). Which approach correctly integrates the digit label condition into the generator model?

hard

A. Add the label as a separate input to the discriminator only

B. Concatenate a one-hot encoded label vector with noise before feeding to the generator

C. Use the label to filter generated images after generation

D. Train the generator without any label input

Solution

1. Step 1: Understand conditional GAN input requirements

   The generator needs both noise and condition (digit label) to create images of specific digits.

2. Step 2: Identify correct integration method

   Concatenating one-hot encoded label with noise vector before input to generator allows conditioning generation on digit.

3. Final Answer:

   Concatenate a one-hot encoded label vector with noise before feeding to the generator -> Option B

4. Quick Check:

   Condition + noise input guides generation [OK]

Hint: Combine label and noise before generator input [OK]

Common Mistakes:

• Conditioning only discriminator, not generator
• Filtering outputs instead of conditioning input
• Ignoring label input in generator