Padding and stride in TensorFlow - Model Metrics & Evaluation

Metrics & Evaluation - Padding and stride

Which metric matters for Padding and Stride and WHY

When using padding and stride in convolutional layers, the key metric to watch is output size and accuracy. Padding controls how the edges of input data are handled, affecting output size and detail preservation. Stride controls how much the filter moves, affecting output size and feature detection. The right balance keeps important features while maintaining good accuracy.

Confusion matrix or equivalent visualization

Input size: 5x5
Filter size: 3x3
Padding: 1 (same padding)
Stride: 1

Output size calculation:
Output = ((Input - Filter + 2*Padding) / Stride) + 1
       = ((5 - 3 + 2*1) / 1) + 1 = 5

If stride = 2:
Output = floor(((5 - 3 + 2*1) / 2)) + 1 = 3

This shows how padding and stride change output size.
    

Precision vs Recall tradeoff (Analogy for Padding and Stride)

Think of padding and stride like zooming in on a photo:

• Padding adds a border so you don't lose edges, like keeping the full photo frame.
• Stride is how big your steps are when scanning the photo.

If stride is too big, you skip details (low recall). If padding is too small, you lose edges (low precision). The tradeoff is to keep enough detail (high recall) without too much noise (high precision).

What "good" vs "bad" metric values look like for Padding and Stride

Good: Output size matches expected dimensions, model accuracy is high, and features are well captured.

Bad: Output size too small or too large, model accuracy drops, or important features are missed due to too large stride or no padding.

Common pitfalls

• Using no padding causes output size to shrink, losing edge information.
• Too large stride skips important features, hurting model learning.
• Incorrect output size calculation leads to shape mismatch errors.
• Ignoring padding and stride effects can cause overfitting or underfitting.

Self-check question

Your convolutional model has 98% accuracy but the output feature map is very small due to stride 3 and no padding. Is this good?

Answer: No. The large stride and no padding shrink the output too much, likely missing important features. Despite high accuracy, the model may not generalize well. Adjust padding and stride to preserve features.

Key Result

Padding and stride control output size and feature capture, directly impacting model accuracy and detail preservation.

Practice

1. What does padding do in a convolutional layer in TensorFlow?

easy

A. Changes the filter size

B. Adjusts the learning rate

C. Controls the number of filters

D. Adds extra space around the input edges

Solution

1. Step 1: Understand padding role

   Padding adds pixels around the input image edges to control output size and preserve edge information.

2. Step 2: Compare options

   Only Adds extra space around the input edges correctly describes padding as adding space around input edges.

3. Final Answer:

   Adds extra space around the input edges -> Option D

4. Quick Check:

   Padding = Adds space around input edges [OK]

Hint: Padding adds border space around input images [OK]

Common Mistakes:

• Confusing padding with stride
• Thinking padding changes filter size
• Assuming padding controls learning rate

2. Which of the following is the correct way to specify stride of 2 in TensorFlow's Conv2D layer?

easy

A. strides=(2)

B. stride=(2)

C. strides=(2, 2)

D. stride=2

Solution

1. Step 1: Recall Conv2D stride syntax

   TensorFlow Conv2D expects strides as a tuple for height and width, e.g., (2, 2).

2. Step 2: Check options

   Only strides=(2, 2) uses the correct parameter name and tuple format.

3. Final Answer:

   strides=(2, 2) -> Option C

4. Quick Check:

   Stride syntax = strides=(2, 2) [OK]

Hint: Use tuple for strides in Conv2D: strides=(height, width) [OK]

Common Mistakes:

• Using 'stride' instead of 'strides'
• Passing a single integer instead of a tuple
• Omitting parentheses around stride values

3. Given an input of size 28x28, a filter size of 3x3, stride of 2, and 'valid' padding, what is the output size after Conv2D?

medium

A. 13x13

B. 14x14

C. 15x15

D. 12x12

Solution

1. Step 1: Calculate output size formula for 'valid' padding

   Output size = floor((Input - Filter) / Stride) + 1 = floor((28 - 3) / 2) + 1 = floor(25 / 2) + 1 = 12 + 1 = 13.

2. Step 2: Confirm output dimensions

   Output height and width are both 13, so output size is 13x13.

3. Final Answer:

   13x13 -> Option A

4. Quick Check:

   Output size = 13x13 with valid padding [OK]

Hint: Use formula: floor((input-filter)/stride)+1 for 'valid' padding [OK]

Common Mistakes:

• Using 'same' padding formula instead
• Forgetting to floor the division
• Adding padding when 'valid' means no padding

4. You wrote this code but get an error:

tf.keras.layers.Conv2D(filters=32, kernel_size=3, strides=2, padding='samee')

What is the error?

medium

A. Strides must be a tuple, not integer

B. Invalid padding value 'samee'

C. Kernel size must be a tuple

D. Filters must be a list

Solution

1. Step 1: Check padding parameter

   Padding accepts only 'valid' or 'same'. 'samee' is a typo and invalid.

2. Step 2: Verify other parameters

   Strides can be integer or tuple, kernel_size can be int or tuple, filters is integer. So no error there.

3. Final Answer:

   Invalid padding value 'samee' -> Option B

4. Quick Check:

   Padding typo causes error [OK]

Hint: Check spelling of padding values: 'same' or 'valid' only [OK]

Common Mistakes:

• Misspelling padding values
• Assuming strides must be tuple always
• Confusing kernel_size type requirements

5. You want to keep the output size same as input (28x28) after Conv2D with filter size 5x5. Which padding and stride should you use?

hard

A. padding='same', strides=(1, 1)

B. padding='valid', strides=(1, 1)

C. padding='same', strides=(2, 2)

D. padding='valid', strides=(2, 2)

Solution

1. Step 1: Understand output size preservation

   To keep output size same as input, use 'same' padding and stride 1 so filter slides over every pixel with padding added.

2. Step 2: Check options

   Only padding='same', strides=(1, 1) uses 'same' padding and stride (1,1), which preserves size.

3. Final Answer:

   padding='same', strides=(1, 1) -> Option A

4. Quick Check:

   Same padding + stride 1 keeps size [OK]

Hint: Use 'same' padding and stride 1 to keep output size [OK]

Common Mistakes:

• Using 'valid' padding which shrinks output
• Using stride > 1 which reduces output size
• Confusing filter size effect on output