Easy

ReLU Activation

Easy

~10 min

code completion

ReLU: Rectified Linear Unit

ReLU is the most popular activation function in deep learning:

ReLU (z) = max (0, z)

It zeroes out negative values and passes positive values unchanged. Key advantages:

Cheap to compute

Does not saturate for positive inputs (no vanishing gradient problem)

Induces sparsity (dead neurons for negative inputs)

In NumPy: np.maximum(0, z)

Note: np.maximum(0, z) is element-wise; np.max(z) takes the global max — different things!

Your task:

Implement relu(z) that applies ReLU element-wise.

Example Tests

Mixed positive and negative values

Input: {"z":[-3,-1,0,1,3]}

Expected: [0,0,0,1,3]

2D array

Input: {"z":[[1,-1],[2,-2]]}

Expected: [[1,0],[2,0]]

All negative: all zeros

Input: {"z":[-5,-4,-3,-2,-1]}

Expected: [0,0,0,0,0]

You can read the full problem statement above. Create a free account to run code in the browser, submit solutions, and track your progress.