Feature extraction strategy in PyTorch - Model Metrics & Evaluation

Metrics & Evaluation - Feature extraction strategy

Which metric matters for Feature Extraction Strategy and WHY

Feature extraction helps the model learn useful information from raw data. The key metrics to check are accuracy or loss on validation data. These show if the extracted features help the model make better predictions.

For classification tasks, accuracy, precision, and recall matter because they tell us how well the features separate classes.

For regression, mean squared error (MSE) or mean absolute error (MAE) show how well features predict continuous values.

In short, metrics that measure prediction quality after feature extraction are most important. They tell if the features are meaningful.

Confusion Matrix Example

Suppose we extract features from images to classify cats vs dogs. After training, the confusion matrix might look like this:

      | Predicted Cat | Predicted Dog |
      |--------------|---------------|
      | True Cat: 50 | False Dog: 10 |
      | False Cat: 5 | True Dog: 35  |
    

Here:

• TP (True Positive) = 50 (correct cat predictions)
• FP (False Positive) = 5 (dog predicted as cat)
• FN (False Negative) = 10 (cat predicted as dog)
• TN (True Negative) = 35 (correct dog predictions)

From this, precision = 50 / (50 + 5) = 0.91, recall = 50 / (50 + 10) = 0.83.

Precision vs Recall Tradeoff with Feature Extraction

Feature extraction affects precision and recall. For example:

• If features are too general, the model may predict many positives, increasing recall but lowering precision (more false alarms).
• If features are too strict, the model predicts fewer positives, increasing precision but lowering recall (misses some true cases).

Example: In medical image analysis, missing a disease (low recall) is worse than false alarms (low precision). So feature extraction should favor recall.

In spam detection, wrongly marking good emails as spam (low precision) is worse, so features should favor precision.

Good vs Bad Metric Values for Feature Extraction

Good feature extraction leads to:

• High accuracy (e.g., > 85% on validation)
• Balanced precision and recall (both > 0.8) for classification
• Low loss values (e.g., cross-entropy loss < 0.5)

Bad feature extraction shows:

• Low accuracy (close to random guessing, e.g., ~50% for binary)
• Very low precision or recall (below 0.5), meaning poor class separation
• High loss values or no improvement during training

Common Pitfalls in Feature Extraction Metrics

• Accuracy paradox: High accuracy can be misleading if classes are imbalanced. Features may ignore minority classes.
• Data leakage: Features accidentally include future or test data info, inflating metrics falsely.
• Overfitting: Features too tuned to training data cause high training accuracy but poor validation results.
• Ignoring metric tradeoffs: Focusing only on accuracy without checking precision/recall can hide poor feature quality.

Self Check

Your model uses extracted features and shows 98% accuracy but only 12% recall on fraud detection. Is it good?

Answer: No. Despite high accuracy, the model misses most fraud cases (low recall). This is bad because catching fraud is critical. Feature extraction should improve recall to detect fraud better.

Key Result

Feature extraction quality is best judged by balanced precision and recall, not just accuracy.

Practice

1. What is the main purpose of using a pre-trained model for feature extraction in PyTorch?

easy

A. To replace the optimizer with a new one

B. To use learned features from a large dataset and avoid training from scratch

C. To train all layers from random weights

D. To increase the size of the dataset automatically

Solution

1. Step 1: Understand feature extraction concept

   Feature extraction uses a model already trained on a large dataset to get useful features without training all layers again.

2. Step 2: Identify the main benefit

   This saves time and resources by reusing learned knowledge instead of starting from scratch.

3. Final Answer:

   To use learned features from a large dataset and avoid training from scratch -> Option B

4. Quick Check:

   Feature extraction = reuse learned features [OK]

Hint: Pre-trained means reuse, not retrain all layers [OK]

Common Mistakes:

• Thinking feature extraction means training all layers
• Confusing feature extraction with data augmentation
• Believing optimizer changes are part of feature extraction

2. Which PyTorch code snippet correctly freezes all layers of a pre-trained model except the final layer?

easy

A. for param in model.parameters(): param.requires_grad = True
model.fc = nn.Linear(512, 10)

B. model.fc.requires_grad = False
for param in model.parameters(): param.requires_grad = True

C. for param in model.parameters(): param.requires_grad = False
model.fc = nn.Linear(512, 10)

D. model.fc = nn.Linear(512, 10)
for param in model.parameters(): param.requires_grad = False

Solution

1. Step 1: Freeze all layers by setting requires_grad to false

   The loop disables gradient updates for all parameters to keep pre-trained weights fixed.

2. Step 2: Replace the final layer with a new one to train

   Assigning a new linear layer to model.fc allows training only this layer for the new task.

3. Final Answer:

   for param in model.parameters(): param.requires_grad = False model.fc = nn.Linear(512, 10) -> Option C

4. Quick Check:

   Freeze all except final layer = for param in model.parameters(): param.requires_grad = False model.fc = nn.Linear(512, 10) [OK]

Hint: Freeze first, then replace final layer [OK]

Common Mistakes:

• Not freezing layers before replacing final layer
• Freezing final layer instead of others
• Setting requires_grad true for all parameters

3. Given this PyTorch code for feature extraction, what will be the output shape of features?

import torch
import torchvision.models as models
model = models.resnet18(pretrained=True)
model.fc = torch.nn.Identity()
input_tensor = torch.randn(4, 3, 224, 224)
features = model(input_tensor)
print(features.shape)

medium

A. torch.Size([4, 512])

B. torch.Size([4, 1000])

C. torch.Size([4, 3, 224, 224])

D. torch.Size([4, 2048])

Solution

1. Step 1: Understand model modification

   Replacing model.fc with Identity removes the final classification layer, so output is the feature vector before classification.

2. Step 2: Know ResNet18 feature size

   ResNet18 outputs a 512-dimensional vector before the final fc layer for each input image.

3. Final Answer:

   torch.Size([4, 512]) -> Option A

4. Quick Check:

   ResNet18 features = 512 dims [OK]

Hint: Identity layer outputs feature vector size [OK]

Common Mistakes:

• Assuming output is 1000 classes without removing fc
• Confusing batch size with feature dimension
• Expecting 2048 features from ResNet18 (it's 512)

4. Identify the error in this feature extraction code snippet and select the fix:

model = models.resnet50(pretrained=True)
for param in model.parameters():
    param.requires_grad = False
model.fc = nn.Linear(2048, 5)
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)

# Training loop here

medium

A. No error; code is correct

B. Set requires_grad=True for model.fc parameters after replacement

C. Use Adam optimizer instead of SGD

D. Remove freezing of parameters to train all layers

Solution

1. Step 1: Check freezing timing

   The loop freezes existing parameters before replacing model.fc, so the new fc layer's parameters are created with requires_grad=True by default.

2. Step 2: Verify optimizer behavior

   Optimizer only updates parameters where requires_grad=True, which are the new fc parameters; backbone remains frozen.

3. Final Answer:

   No error; code is correct -> Option A

4. Quick Check:

   New layer params unfrozen by default [OK]

Hint: New layers have requires_grad=True by default [OK]

Common Mistakes:

• Assuming freezing all parameters includes new layers
• Changing optimizer without fixing requires_grad
• Removing freezing unnecessarily

5. You want to use a pre-trained ResNet34 to classify 3 classes in your dataset. You freeze all layers except the last one. However, your training accuracy stays very low. What is the best next step to improve feature extraction performance?

hard

A. Reduce batch size to 1 to improve gradient estimates

B. Increase learning rate to 1.0 for faster training

C. Replace the optimizer with SGD without momentum

D. Unfreeze some deeper layers to fine-tune features for your task

Solution

1. Step 1: Understand freezing impact

   Freezing all but last layer may limit model's ability to adapt features to new classes, causing low accuracy.

2. Step 2: Fine-tune some deeper layers

   Unfreezing some layers closer to output allows the model to adjust features better for your specific dataset.

3. Final Answer:

   Unfreeze some deeper layers to fine-tune features for your task -> Option D

4. Quick Check:

   Fine-tune layers = better adaptation [OK]

Hint: Fine-tune layers if frozen model underperforms [OK]

Common Mistakes:

• Increasing learning rate too much causes instability
• Changing optimizer without addressing feature adaptation
• Reducing batch size unnecessarily