Activation function leaky relu

184. Activation function leaky relu

easy

General

senior

Implement the Leaky ReLU activation function, a small twist on ReLU that keeps a tiny gradient for negative inputs to help training stay stable. You’ll apply it element-wise to a vector of inputs from a basic deep learning forward pass.

f(x) = \begin{cases} x & \text{if } x \ge 0 \\ \alpha x & \text{if } x < 0 \end{cases}

Requirements

Implement the function

python

Rules:

Apply Leaky ReLU element-wise using: $f(x) = \max(x, \alpha x)$
Use only NumPy and Python built-in libraries (no deep learning frameworks).
Do not use any prebuilt activation helpers (e.g., from PyTorch/TensorFlow).
Keep the function as a single forward-pass utility (no backprop needed).

Example

python

Output:

python

Input Signature

Argument	Type
x	np.ndarray
alpha	float

Output Signature

Return Name	Type
value	np.ndarray

Constraints

Return NumPy array
Element-wise on 1D array only
Use only NumPy + built-ins

Hint 1

Start from the definition: for each scalar v, output max(v, alpha*v).

Hint 2

Translate the scalar rule into an element-wise operation over the 1D input using np.where.

Hint 3

Handle negatives with alpha*v while leaving v >= 0 unchanged.

Roles

ML Engineer

AI Engineer

Companies

General

Levels

senior

entry

Input Arguments

Edit values below to test with custom inputs

You need tolog in/sign upto run or submit