0% found this document useful (0 votes)
5 views6 pages

Module 1 & Module 2 Viva Questions

The document covers key concepts related to Transformer models and Generative Adversarial Networks (GANs), including their architectures, training processes, and components like tokenizers and attention mechanisms. It explains the advantages of Transformers over RNNs, the roles of Generator and Discriminator in GANs, and the use of diffusion models for image generation. Additionally, it discusses the importance of parameters in training and model performance, as well as the evaluation of generated outputs.

Uploaded by

Raj Singh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
5 views6 pages

Module 1 & Module 2 Viva Questions

The document covers key concepts related to Transformer models and Generative Adversarial Networks (GANs), including their architectures, training processes, and components like tokenizers and attention mechanisms. It explains the advantages of Transformers over RNNs, the roles of Generator and Discriminator in GANs, and the use of diffusion models for image generation. Additionally, it discusses the importance of parameters in training and model performance, as well as the evaluation of generated outputs.

Uploaded by

Raj Singh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Module 1

Viva Questions
1. What is a Transformer model, and why is it preferred over RNNs for text classification?
A Transformer is a deep learning model based entirely on the attention mechanism,
without using recurrence or convolution.
It is preferred over RNNs because:
 It processes tokens in parallel, making training faster
 It captures long-range dependencies more effectively
 It avoids issues like vanishing gradients
 It scales well to large datasets

2. Which Hugging Face classes are typically used to build a Transformer-based text
classification model?
Commonly used Hugging Face classes include:
 AutoTokenizer – for text tokenization
 AutoModelForSequenceClassification – for classification models
 Trainer – to handle training and evaluation
 TrainingArguments – to configure training parameters
 pipeline – for quick inference

3. What is the role of the tokenizer in the Transformer pipeline?


The tokenizer:
 Converts raw text into tokens
 Maps tokens to numerical IDs
 Adds special tokens like [CLS] and [SEP]
 Creates attention masks
Transformers only understand numbers, so tokenization is essential.

4. What happens if text is not tokenized properly?


If tokenization is incorrect:
 The model may misinterpret words
 Sentence structure can be lost
 Important context may be removed
 Model accuracy decreases significantly
Incorrect tokenization leads to poor or meaningless predictions.

5. What is role of transformers, torch and pipeline in python code? What are these?
Transformers provides pre-trained models, tokenizers, and training utilities from Hugging
Face.
torch (PyTorch) handles tensor operations, model training, backpropagation, and GPU
acceleration.

Pipeline:- A high-level API for quick tasks like text classification without writing full
training code.

6. What is the purpose of attention masks in text classification models?


Attention masks:
 Indicate which tokens are real and which are padding
 Prevent the model from attending to padding tokens
 Improve prediction accuracy
They ensure the model focuses only on meaningful text.

7. What is the difference between fine-tuning a Transformer and training a model from
scratch?

8. Fine-Tuning Training from Scratch


Uses pre-trained weights Random weight initialization
Requires less data Needs large datasets
Faster training Slower and expensive
Better performance Harder to converge

Fine-tuning is preferred for most NLP tasks.

9. How does batch size affect training stability and GPU memory usage in your
implementation?

Large batch size

a. More stable gradients


b. Faster training
c. Requires more GPU memory

Small batch size

d. Noisier gradients
e. Slower training
f. Lower memory usage

Batch size is chosen based on GPU capacity and model size.

10. What is the attention mechanism, and why is it important in Transformer architectures?
The attention mechanism allows the model to:
 Focus on important words in a sentence
 Assign different importance weights to tokens
 Capture contextual meaning efficiently
It replaces recurrence and enables parallel processing

11. Explain the difference between Query (Q), Key (K), and Value (V) in attention
computation.
Query (Q): What the model is looking for
Key (K): What each word offers
Value (V): Actual information passed forward
Attention scores measure how well Q matches K, and then weight the V vectors.

12. How is the attention score calculated mathematically in self-attention?

Where:
Q K computes similarity
T

 √ d k stabilizes gradients
 Softmax converts scores to probabilities

13. What does an attention heatmap represent in visualization?


An attention heatmap:
 Visualizes attention weights between tokens
 Shows which words influence others
 Darker colors indicate higher attention
It helps in model interpretability and understanding predictions

Module 2
Viva Questions
1. What is a GAN, and which two neural networks are implemented in your DeepFake
code?
A GAN consists of a Generator and a Discriminator trained together to generate
realistic fake data.
2. What role does the Generator play in a DeepFake application?
The Generator creates synthetic face images that try to look real.
3. How does it learn to generate realistic faces?
The Generator learns to generate realistic faces through adversarial training. It starts by
producing random images from noise. These images are evaluated by the Discriminator,
which provides a loss signal indicating how fake or real the images appear. Using this
loss, the Generator updates its weights via backpropagation to reduce the difference
between generated and real images. Over many iterations, the Generator learns facial
features such as eyes, nose, texture, and expressions, gradually producing realistic faces.

4. What is the function of the Discriminator, and how does it help improve the Generator?
The Discriminator acts as a binary classifier that distinguishes between real images from
the dataset and fake images generated by the Generator. It outputs a probability score
indicating whether an image is real or fake. When the Discriminator correctly classifies
images, it produces meaningful gradients that are backpropagated not only to update
itself but also passed to the Generator. This feedback helps the Generator understand
which features look unrealistic, forcing it to improve image quality over time.

5. Which loss functions are used for the Generator and Discriminator in your code, and
why?
In standard GAN implementations, Binary Cross-Entropy (BCE) loss is used for both
networks:
Discriminator Loss
The Discriminator is trained to:
 Assign label 1 to real images
 Assign label 0 to fake images
This loss measures how well the Discriminator separates real and fake samples.

Generator Loss
The Generator is trained to:
 Make fake images be classified as real (label = 1)

6. How do you evaluate the quality of DeepFake outputs generated by your GAN?
The quality of DeepFake outputs is evaluated using a combination of:
 Visual inspection to assess realism, facial consistency, and artifacts
 Discriminator accuracy/loss to check how well fake images resemble real ones

7. Why diffusers are used? Which diffuser model you have imported in your python code?
How do you import the diffuser model in your code?
Diffusers are used because they:
 Produce high-quality and diverse images
 Provide stable training and inference
 Offer pretrained diffusion pipelines
In the code, a commonly used model is Stable Diffusion.
from diffusers import StableDiffusionPipeline
import torch

device = "cuda" if [Link].is_available() else "cpu"

pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
torch_dtype=torch.float16
).to(device)

prompt = "a realistic photo of a modern house with a garden, daylight, high detail"

image = pipe(prompt, num_inference_steps=30, guidance_scale=7.5).images[0]

[Link]("sd_output.png")
[Link]()

This pipeline includes the UNet, scheduler, and text encoder required for image
generation.

8. What is a diffusion model, and how does it differ from GAN-based image generation?
A diffusion model generates images by gradually removing noise from a random signal,
whereas GANs generate images in a single forward pass.

Diffusion Model GAN


Step-by-step generation One-shot generation
Stable training Unstable adversarial training
High diversity Risk of mode collapse
Slower inference Faster inference

9. Explain the forward diffusion process implemented in your code.


The forward diffusion process:
 Gradually adds Gaussian noise to training images
 Continues over multiple timesteps until the image becomes pure noise
 Is a fixed, non-learnable process
This prepares the data for training the reverse (denoising) model.

10. What is the reverse diffusion process, and how does the model learn it?
The reverse diffusion process:
 Gradually removes noise from a noisy image
 Uses a neural network to predict the noise at each timestep
 Is learned through supervised training using noisy-clean image pairs
The model learns to reconstruct images by minimizing noise prediction loss.

11. Why is Gaussian noise added gradually during the diffusion process instead of all at
once?
Gradual noise addition:
 Makes the reverse process easier to learn
 Maintains a smooth probability transition
 Prevents loss of structural information early
Adding all noise at once would make reconstruction extremely difficult.

12. What is the role of the UNet architecture in diffusion models?


The UNet:
 Predicts the noise component at each timestep
 Uses encoder–decoder structure with skip connections
 Preserves fine-grained spatial details during denoising
It is the core denoising network in diffusion models.

13. How do parameters like num_inference_steps and guidance_scale affect image quality
and generation speed?

num_inference_steps

a. More steps → better quality, slower generation


b. Fewer steps → faster but lower quality

guidance_scale

c. Higher value → stronger prompt adherence, less diversity


d. Lower value → more diverse but less controlled images

14. Why are diffusion models considered more stable to train compared to GANs?
Diffusion models are more stable because:
 They avoid adversarial training
 Use a simple mean squared error (MSE) objective
 Have well-defined likelihood-based learning
 Do not suffer from mode collapse
This results in predictable and reliable training.

You might also like