Feature extraction approach in TensorFlow - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of feature extraction with TensorFlow Hub

What is the shape of the output tensor after extracting features from a batch of 3 images using a TensorFlow Hub MobileNetV2 feature vector model?

TensorFlow

import tensorflow as tf
import tensorflow_hub as hub

model_url = "https://tfhub.dev/google/tf2-preview/mobilenet_v2/feature_vector/4"
feature_extractor = hub.KerasLayer(model_url, input_shape=(224, 224, 3))

batch_images = tf.random.uniform([3, 224, 224, 3])
features = feature_extractor(batch_images)
output_shape = features.shape
print(output_shape)

A (3, 1280)

B (224, 224, 3)

C (3, 1001)

D (1, 1280)

💡 Hint

The MobileNetV2 feature vector outputs a 1280-dimensional vector per image.

The feature extractor outputs a 2D tensor where each row corresponds to one image's feature vector. Since the batch size is 3, the output shape is (3, 1280).

❓ Model Choice

intermediate

Choosing a pre-trained model for feature extraction

You want to extract features from images to use in a new classifier. Which pre-trained model is best suited for feature extraction without the classification head?

A A fully connected network trained on tabular data

B ResNet50 with include_top=True

C ResNet50 with include_top=False

D A randomly initialized CNN

💡 Hint

Feature extraction requires a model trained on images without the final classification layer.

Using ResNet50 with include_top=False removes the classification head, allowing extraction of image features useful for new tasks.

❓ Hyperparameter

advanced

Effect of freezing layers during feature extraction

When using a pre-trained CNN for feature extraction, what is the effect of freezing all convolutional layers during training of a new classifier on top?

A The convolutional layers' weights do not update, speeding up training and preserving learned features.

B The convolutional layers' weights update, allowing the model to learn new features from scratch.

C The model will overfit immediately due to frozen layers.

D The model will not train because frozen layers cause errors.

💡 Hint

Freezing layers means their weights stay fixed during training.

Freezing convolutional layers keeps their learned features intact and reduces training time since fewer parameters update.

❓ Metrics

advanced

Evaluating feature extraction effectiveness

You extracted features from images using a pre-trained model and trained a new classifier. Which metric best indicates if the extracted features are useful for the classification task?

A Training loss of the pre-trained model

B Validation accuracy of the new classifier

C Number of layers in the pre-trained model

D Size of the input images

💡 Hint

The quality of features is reflected in how well the classifier performs on unseen data.

Validation accuracy shows how well the classifier generalizes using the extracted features, indicating their usefulness.

🔧 Debug

expert

Debugging feature extraction output shape mismatch

You use a TensorFlow Hub feature extractor expecting output shape (None, 2048), but your code raises a shape mismatch error. Which code change fixes this?

TensorFlow

import tensorflow_hub as hub
import tensorflow as tf

feature_extractor = hub.KerasLayer('https://tfhub.dev/google/imagenet/resnet_v2_50/feature_vector/5', input_shape=(224, 224, 3))

inputs = tf.random.uniform([5, 224, 224, 3])
features = feature_extractor(inputs)
print(features.shape)

A Change input_shape to (None, 224, 224, 3) in KerasLayer

B Resize input images to (299, 299, 3) before passing to feature_extractor

C Use a different model URL that outputs shape (None, 1280)

D Change input_shape to (224, 224, 3) and ensure inputs have batch size dimension

💡 Hint

The input_shape parameter expects the shape of a single image, excluding batch size.

The input_shape should be (224, 224, 3) for single images; inputs must have batch dimension. This matches the model's expected input and output shapes.

Practice

1. What is the main purpose of using feature extraction with a pre-trained TensorFlow model?

easy

A. To use learned features from a large dataset to help a new task

B. To train a model from scratch with random weights

C. To increase the size of the dataset automatically

D. To reduce the number of layers in the model

Solution

1. Step 1: Understand feature extraction concept

   Feature extraction uses a pre-trained model to get useful features learned from a large dataset.

2. Step 2: Identify the benefit for new tasks

   This helps when you have little data or want faster training by reusing learned features.

3. Final Answer:

   To use learned features from a large dataset to help a new task -> Option A

4. Quick Check:

   Feature extraction = reuse learned features [OK]

Hint: Feature extraction reuses knowledge from big datasets [OK]

Common Mistakes:

• Thinking feature extraction trains from scratch
• Confusing feature extraction with data augmentation
• Believing it reduces model layers automatically

2. Which TensorFlow code snippet correctly freezes the base model layers for feature extraction?

easy

A. base_model.trainable = True

B. base_model.trainable = False

C. base_model.freeze()

D. base_model.layers.freeze = True

Solution

1. Step 1: Recall how to freeze layers in TensorFlow

   Setting trainable = False on the base model freezes its weights.

2. Step 2: Check each option's correctness

   base_model.trainable = False correctly sets base_model.trainable = False. Others are invalid syntax or do not freeze layers.

3. Final Answer:

   base_model.trainable = False -> Option B

4. Quick Check:

   Freeze layers = trainable False [OK]

Hint: Freeze layers by setting trainable to False [OK]

Common Mistakes:

• Using trainable = True to freeze layers
• Calling non-existent freeze() method
• Trying to set freeze attribute on layers directly

3. Given this code, what is the output shape of features?

base_model = tf.keras.applications.MobileNetV2(input_shape=(128,128,3), include_top=False)
base_model.trainable = False
inputs = tf.keras.Input(shape=(128,128,3))
features = base_model(inputs)
print(features.shape)

medium

A. (None, 4, 4, 1280)

B. (None, 7, 7, 512)

C. (None, 1, 1, 1000)

D. (None, 128, 128, 3)

Solution

1. Step 1: Understand MobileNetV2 output with include_top=False

   MobileNetV2 outputs feature maps of size (4,4,1280) for 128x128 input when top layers are excluded.

2. Step 2: Confirm batch dimension is None

   The batch size is None (unspecified), so output shape is (None, 4, 4, 1280).

3. Final Answer:

   (None, 4, 4, 1280) -> Option A

4. Quick Check:

   MobileNetV2 feature shape = (None,4,4,1280) [OK]

Hint: MobileNetV2 outputs 4x4x1280 features for 128x128 input [OK]

Common Mistakes:

• Confusing output with classification layer shape
• Assuming output shape equals input shape
• Ignoring batch dimension None

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

base_model = tf.keras.applications.ResNet50(weights='imagenet', include_top=False)
base_model.trainable = False
inputs = tf.keras.Input(shape=(224,224,3))
x = base_model(inputs)
x = tf.keras.layers.Dense(10)(x)
model = tf.keras.Model(inputs, x)

medium

A. Base model weights not loaded correctly

B. Input shape is incorrect for ResNet50

C. Dense layer applied to 4D tensor without flattening

D. trainable should be True to freeze layers

Solution

1. Step 1: Check tensor shapes before Dense layer

   Base model outputs 4D tensor (batch, height, width, channels). Dense expects 2D input.

2. Step 2: Identify missing flatten or global pooling

   Applying Dense directly to 4D tensor causes error; flatten or global pooling needed first.

3. Final Answer:

   Dense layer applied to 4D tensor without flattening -> Option C

4. Quick Check:

   Dense needs 2D input, flatten first [OK]

Hint: Flatten or pool before Dense layer [OK]

Common Mistakes:

• Applying Dense directly to 4D output
• Confusing input shape requirements
• Misunderstanding trainable flag usage

5. You want to use feature extraction with MobileNetV2 on 64x64 images but get an error when adding a Dense layer after the base model. What is the best way to fix this?

hard

A. Set base_model.trainable = True to allow training

B. Change input shape to 224x224 to match MobileNetV2 default

C. Remove the Dense layer and use base model output directly

D. Add a GlobalAveragePooling2D layer before the Dense layer

Solution

1. Step 1: Understand why Dense layer causes error

   MobileNetV2 outputs 4D feature maps; Dense layers need 2D input.

2. Step 2: Use GlobalAveragePooling2D to convert 4D to 2D

   Adding GlobalAveragePooling2D flattens spatial dimensions, making output suitable for Dense.

3. Step 3: Evaluate other options

   Changing input size is not required; training base model is optional; removing Dense loses classification ability.

4. Final Answer:

   Add a GlobalAveragePooling2D layer before the Dense layer -> Option D

5. Quick Check:

   Pool features before Dense layer [OK]

Hint: Use pooling to flatten features before Dense [OK]

Common Mistakes:

• Changing input size unnecessarily
• Trying to train frozen base model
• Skipping pooling and causing shape errors