Sigmoid Activation Function
89.2% acceptance
EasyNeural Networks
Commonly asked at:
GoogleWhy this matters:
Tests knowledge of activation functions – essential for understanding neural network forward/backward passes. Sigmoid is particularly useful in binary classification tasks.
The sigmoid function maps any real number to the range (0, 1). You need to implement both the forward pass (sigmoid) and the backward pass (sigmoid derivative) for use in neural network training.
Examples
Input:
x = 0
Output:
sigmoid(0) = 0.5
Explanation:
When x=0, e^0=1, so sigmoid(0) = 1/(1+1) = 0.5
Constraints
- Handle numerical stability for large |x| values
- Input can be a scalar or numpy array
- Output should match input shape
Tags
transformersmodel-evaluationclassic-nlpclassification