Fine-tuning approach in Computer Vision - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the fine-tuning approach in machine learning?

Fine-tuning is a method where a pre-trained model is adapted to a new, related task by continuing training on new data. It helps the model learn specific features of the new task while keeping general knowledge from before.

beginner

Why do we use fine-tuning instead of training a model from scratch?

Fine-tuning saves time and resources because it starts from a model that already knows useful features. It also often improves performance, especially when the new dataset is small.

intermediate

What does it mean to 'freeze' layers during fine-tuning?

Freezing layers means keeping some parts of the pre-trained model fixed so they don't change during training. This helps keep the general knowledge intact while only adjusting other parts for the new task.

intermediate

How does fine-tuning help in computer vision tasks?

In computer vision, fine-tuning allows models trained on large image datasets to adapt to specific tasks like recognizing new objects or styles, improving accuracy without needing huge new datasets.

intermediate

What is a common strategy for choosing which layers to fine-tune?

A common strategy is to freeze early layers that capture basic features like edges and textures, and fine-tune later layers that learn task-specific details.

What is the main benefit of fine-tuning a pre-trained model?

A It guarantees perfect accuracy

B It removes the need for any new data

C It trains the model from scratch

D It reduces training time and improves performance on new tasks

Fine-tuning uses existing knowledge to quickly adapt to new tasks, saving time and often improving results.

During fine-tuning, what does freezing layers do?

A Prevents some layers from updating weights

B Deletes those layers

C Adds new layers to the model

D Increases the learning rate

Freezing layers keeps their weights fixed so they don't change during training.

Which layers are usually fine-tuned in computer vision models?

A Early layers only

B Later layers that capture task-specific features

C Only the input layer

D All layers equally

Later layers learn details specific to the new task, so they are often fine-tuned.

Fine-tuning is especially useful when the new dataset is:

A Very small

B Very large

C Unrelated to the original task

D Perfectly labeled

Fine-tuning helps when new data is limited by leveraging prior knowledge.

What is a pre-trained model?

A A model with no training

B A model trained only on the new task

C A model trained on a large dataset before fine-tuning

D A model that cannot be fine-tuned

Pre-trained models have learned general features from large datasets and can be adapted to new tasks.

Explain the fine-tuning approach and why it is useful in computer vision.

Think about how a model trained on many images can help with a new image task.

Describe the role of freezing layers during fine-tuning and how to decide which layers to freeze.

Consider which parts of the model learn basics vs. details.

Practice

1. What is the main purpose of fine-tuning a pre-trained computer vision model?

easy

A. To adapt the model to a new task using less data and time

B. To train a model from scratch with a large dataset

C. To increase the size of the model for better accuracy

D. To remove layers from the model to make it smaller

Solution

1. Step 1: Understand fine-tuning concept

   Fine-tuning means starting from a model already trained on a related task.

2. Step 2: Identify the benefit

   This approach saves time and data by reusing learned features for a new task.

3. Final Answer:

   To adapt the model to a new task using less data and time -> Option A

4. Quick Check:

   Fine-tuning = adapt pre-trained model fast [OK]

Hint: Fine-tuning means reusing a model to learn new tasks faster [OK]

Common Mistakes:

• Thinking fine-tuning trains from scratch
• Assuming fine-tuning always increases model size
• Confusing fine-tuning with pruning layers

2. Which code snippet correctly freezes the layers of a PyTorch model before fine-tuning?

easy

A. for param in model.parameters(): param.requires_grad = False

B. model.freeze_layers()

C. model.trainable = False

D. for layer in model.layers: layer.trainable = True

Solution

1. Step 1: Recall PyTorch freezing syntax

   In PyTorch, freezing means setting requires_grad = False for parameters.

2. Step 2: Match code to syntax

   for param in model.parameters(): param.requires_grad = False correctly loops over parameters and disables gradient updates.

3. Final Answer:

   for param in model.parameters(): param.requires_grad = False -> Option A

4. Quick Check:

   Freeze layers = requires_grad False [OK]

Hint: Freeze layers by setting requires_grad = False in PyTorch [OK]

Common Mistakes:

• Using non-existent methods like freeze_layers()
• Setting model.trainable instead of parameters
• Confusing trainable True/False for freezing

3. Given this PyTorch code snippet for fine-tuning, what will be the output of print(sum(p.requires_grad for p in model.parameters())) after freezing layers?

medium

A. Raises an error

B. Number of all model parameters

C. 0

D. Number of unfrozen parameters

Solution

1. Step 1: Understand freezing effect on requires_grad

   Freezing sets requires_grad = False for all parameters.

2. Step 2: Calculate sum of requires_grad flags

   Since all are False, sum counts zero True values.

3. Final Answer:

   0 -> Option C

4. Quick Check:

   All frozen means requires_grad sum = 0 [OK]

Hint: Frozen layers have requires_grad = False, sum is zero [OK]

Common Mistakes:

• Assuming sum counts total parameters
• Thinking sum counts unfrozen parameters without freezing
• Expecting an error from requires_grad attribute

4. You tried fine-tuning but the model's accuracy did not improve. Which mistake could cause this?

medium

A. Using a pre-trained model instead of training from scratch

B. Freezing all layers and not unfreezing any

C. Adding more layers without training them

D. Using a very high learning rate during fine-tuning

Solution

1. Step 1: Identify learning rate impact

   A very high learning rate can cause unstable training and no improvement.

2. Step 2: Evaluate other options

   Freezing all layers prevents learning but usually keeps baseline accuracy; pre-trained models help; adding untrained layers alone doesn't prevent improvement if trained.

3. Final Answer:

   Using a very high learning rate during fine-tuning -> Option D

4. Quick Check:

   High learning rate = no improvement [OK]

Hint: Use smaller learning rates for fine-tuning to improve accuracy [OK]

Common Mistakes:

• Ignoring learning rate effects
• Assuming freezing all layers always improves
• Thinking training from scratch is better always

5. You want to fine-tune a pre-trained CNN for a new image classification task with 5 classes. Which sequence of steps is best practice?

hard

A. Train entire model from scratch with random weights for 5 classes

B. Freeze all layers, replace final layer with 5 outputs, train only final layer, then unfreeze some layers and fine-tune with low learning rate

C. Replace final layer with 5 outputs and train all layers at once with high learning rate

D. Freeze final layer, train earlier layers only, then unfreeze final layer

Solution

1. Step 1: Replace final layer for new classes

   Adjust output layer to match 5 classes for the new task.

2. Step 2: Freeze old layers and train new layer first

   This preserves learned features and trains new output layer quickly.

3. Step 3: Unfreeze some layers and fine-tune with low learning rate

   This improves model performance by adapting features carefully without large updates.

4. Final Answer:

   Freeze all layers, replace final layer with 5 outputs, train only final layer, then unfreeze some layers and fine-tune with low learning rate -> Option B

5. Quick Check:

   Stepwise fine-tuning with low LR = best practice [OK]

Hint: Freeze, replace output, train new layer, then unfreeze and fine-tune [OK]

Common Mistakes:

• Training all layers at once with high learning rate
• Training from scratch ignoring pre-trained weights
• Freezing final layer instead of earlier layers