GAN training loop in PyTorch - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of discriminator loss calculation in GAN training loop

Consider the following snippet from a GAN training loop where the discriminator loss is calculated. What is the output of print(d_loss.item()) after one backward pass?

PyTorch

import torch
import torch.nn as nn

criterion = nn.BCELoss()
d_real = torch.tensor([0.9], requires_grad=True)
d_fake = torch.tensor([0.1], requires_grad=True)
real_labels = torch.ones_like(d_real)
fake_labels = torch.zeros_like(d_fake)

loss_real = criterion(d_real, real_labels)
loss_fake = criterion(d_fake, fake_labels)
d_loss = loss_real + loss_fake
d_loss.backward()
print(round(d_loss.item(), 3))

A 0.211

B 0.000

C 0.693

D 0.105

💡 Hint

Recall that BCELoss for a prediction close to the true label is low, and for wrong predictions is high.

The BCELoss for d_real=0.9 with label 1 is about 0.105, and for d_fake=0.1 with label 0 is about 0.105. Their sum is approximately 0.211.

❓ Model Choice

intermediate

Choosing the correct optimizer update step in GAN training

In a GAN training loop, after computing the discriminator loss and calling d_loss.backward(), which of the following is the correct next step to update the discriminator's weights?

A loss.backward()

B optimizer_d.zero_grad()

C optimizer_d.step()

D optimizer_g.step()

💡 Hint

Think about which optimizer corresponds to the discriminator.

After backward(), calling optimizer_d.step() updates the discriminator weights. zero_grad() clears gradients and should be called before backward().

❓ Hyperparameter

advanced

Effect of learning rate on GAN training stability

Which of the following learning rate choices is most likely to cause unstable GAN training with mode collapse?

A Learning rate = 0.0005

B Learning rate = 0.0002

C Learning rate = 0.00005

D Learning rate = 0.1

💡 Hint

High learning rates can cause large weight updates and instability.

A learning rate of 0.1 is too high for GANs and often causes unstable training and mode collapse.

🔧 Debug

advanced

Identifying the bug in GAN generator training step

In the generator training step below, which line causes the generator to not learn properly?

PyTorch

optimizer_g.zero_grad()
fake_data = generator(noise)
d_fake = discriminator(fake_data)
g_loss = criterion(d_fake, real_labels)
g_loss.backward()
optimizer_d.step()  # <-- suspicious line

A g_loss.backward()

B optimizer_d.step()

C optimizer_g.zero_grad()

D criterion(d_fake, real_labels)

💡 Hint

Check which optimizer should update the generator weights.

Calling optimizer_d.step() updates discriminator weights instead of generator weights, so generator does not learn.

🧠 Conceptual

expert

Why alternate training of discriminator and generator is important in GANs

Why do GAN training loops alternate between updating the discriminator and the generator instead of updating both simultaneously?

A To prevent one network from overpowering the other, maintaining a balance for stable training

B Because simultaneous updates cause syntax errors in PyTorch

C To reduce the total training time by skipping some updates

D Because the generator requires more frequent updates than the discriminator

💡 Hint

Think about the adversarial nature of GANs and training stability.

Alternating updates keep discriminator and generator balanced, preventing one from dominating and causing unstable training.

Practice

1. In a GAN training loop, what is the main role of the discriminator?

easy

A. To calculate the learning rate

B. To generate new fake data samples

C. To distinguish between real and fake data

D. To optimize the generator's weights directly

Solution

1. Step 1: Understand the discriminator's purpose

   The discriminator's job is to tell apart real data from fake data created by the generator.

2. Step 2: Differentiate roles in GAN

   The generator creates fake data, while the discriminator judges it. It does not generate data or adjust learning rates.

3. Final Answer:

   To distinguish between real and fake data -> Option C

4. Quick Check:

   Discriminator = spot fake data [OK]

Hint: Discriminator judges data authenticity, generator creates data [OK]

Common Mistakes:

• Confusing discriminator with generator
• Thinking discriminator generates data
• Assuming discriminator sets learning rate

2. Which of the following is the correct way to zero the gradients of both generator and discriminator optimizers in PyTorch during a GAN training loop?

easy

A. optimizer_G.clear(); optimizer_D.clear()

B. optimizer_G.zero_grad(); optimizer_D.zero_grad()

C. optimizer_G.reset_grad(); optimizer_D.reset_grad()

D. optimizer_G.zero(); optimizer_D.zero()

Solution

1. Step 1: Recall PyTorch optimizer method

   In PyTorch, the method to clear gradients before backpropagation is zero_grad().

2. Step 2: Check each option's validity

   Only optimizer_G.zero_grad(); optimizer_D.zero_grad() uses zero_grad(), which is the correct method. Other options use non-existent methods.

3. Final Answer:

   optimizer_G.zero_grad(); optimizer_D.zero_grad() -> Option B

4. Quick Check:

   Clear gradients = zero_grad() [OK]

Hint: Use zero_grad() to clear gradients before backward pass [OK]

Common Mistakes:

• Using non-existent optimizer methods
• Forgetting to zero gradients before backward
• Confusing zero_grad() with optimizer step()

3. Given the following PyTorch GAN training snippet, what will be the printed output for print(loss_D.item(), loss_G.item()) after one iteration?

real_data = torch.ones(4)
fake_data = torch.zeros(4)
criterion = torch.nn.BCELoss()

# Discriminator output
output_real = torch.tensor([0.9, 0.8, 0.95, 0.85])
output_fake = torch.tensor([0.1, 0.2, 0.05, 0.15])

loss_real = criterion(output_real, real_data)
loss_fake = criterion(output_fake, fake_data)
loss_D = (loss_real + loss_fake) / 2

# Generator tries to fool discriminator
output_fake_for_G = torch.tensor([0.4, 0.3, 0.5, 0.6])
loss_G = criterion(output_fake_for_G, real_data)

print(loss_D.item(), loss_G.item())

medium

A. Approximately 0.693 and 0.105

B. Approximately 0.0 and 1.0

C. Approximately 0.5 and 0.5

D. Approximately 0.136 and 0.831

Solution

1. Step 1: Calculate discriminator loss

   Loss_real = BCELoss([0.9,0.8,0.95,0.85], [1,1,1,1]) ≈ 0.136; Loss_fake = BCELoss([0.1,0.2,0.05,0.15], [0,0,0,0]) ≈ 0.136; Average loss_D ≈ 0.136.

2. Step 2: Calculate generator loss

   Generator tries to fool discriminator, so loss_G = BCELoss([0.4,0.3,0.5,0.6], [1,1,1,1]) ≈ 0.831.

3. Final Answer:

   Approximately 0.136 and 0.831 -> Option D

4. Quick Check:

   Discriminator loss < Generator loss [OK]

Hint: Discriminator loss low near 0, generator loss higher when fooled less [OK]

Common Mistakes:

• Mixing up real and fake labels in loss
• Confusing which loss belongs to generator
• Ignoring averaging of discriminator losses

4. Identify the error in this GAN training loop snippet and select the fix:

for real_data in dataloader:
    optimizer_D.zero_grad()
    output_real = discriminator(real_data)
    loss_real = criterion(output_real, torch.ones_like(output_real))

    fake_data = generator(torch.randn(batch_size, noise_dim))
    output_fake = discriminator(fake_data.detach())
    loss_fake = criterion(output_fake, torch.zeros_like(output_fake))

    loss_D = loss_real + loss_fake
    loss_D.backward()
    optimizer_D.step()

    optimizer_G.zero_grad()
    output_fake_for_G = discriminator(fake_data)
    loss_G = criterion(output_fake_for_G, torch.ones_like(output_fake_for_G))
    loss_G.backward()
    optimizer_D.step()

medium

A. Replace last optimizer_D.step() with optimizer_G.step()

B. Add optimizer_G.zero_grad() before loss_G.backward()

C. Detach fake_data before passing to generator

D. Use torch.zeros_like for output_fake_for_G labels

Solution

1. Step 1: Analyze optimizer steps

   After computing generator loss and backward, the generator's optimizer should update weights, not discriminator's.

2. Step 2: Identify incorrect optimizer call

   The code calls optimizer_D.step() twice; the second should be optimizer_G.step() to update generator.

3. Final Answer:

   Replace last optimizer_D.step() with optimizer_G.step() -> Option A

4. Quick Check:

   Generator update uses optimizer_G.step() [OK]

Hint: Generator updates need optimizer_G.step(), not optimizer_D.step() [OK]

Common Mistakes:

• Calling discriminator optimizer twice
• Forgetting to update generator weights
• Misplacing zero_grad calls

5. You want to improve GAN training stability by modifying the training loop. Which change below best helps prevent the discriminator from overpowering the generator?

hard

A. Train the discriminator less frequently than the generator

B. Increase the learning rate of the discriminator only

C. Remove the generator's loss calculation

D. Use the same optimizer instance for both models

Solution

1. Step 1: Understand discriminator overpowering

   If the discriminator learns too fast, the generator receives poor feedback and training stalls.

2. Step 2: Evaluate options to balance training

   Training discriminator less often than generator slows its learning, giving generator a chance to improve.

3. Final Answer:

   Train the discriminator less frequently than the generator -> Option A

4. Quick Check:

   Balance training frequency to stabilize GAN [OK]

Hint: Train discriminator less often to balance GAN training [OK]

Common Mistakes:

• Increasing discriminator learning rate worsens imbalance
• Removing generator loss stops generator learning
• Sharing optimizer causes parameter update conflicts