HardNLPnlp-372

Layer Normalized Embedding Vector

Problem

Layer normalization centers and scales a vector: (x - mean) / sqrt(variance + eps) using the population variance.

Given the vector and eps, return the normalized vector with gamma = 1 and beta = 0.

Hint

Center first, then divide by the root variance.

case 1

in: [[1,2,3],0.000001]

expected: [-1.224743952833969,0,1.224743952833969]

case 2

in: [[0,0],0.000001]

expected: [0,0]

case 3

in: [[4,4,4,4],0.000001]

expected: [0,0,0,0]

Starter codePython

def layer_normalized_embedding_vector(x, eps):
    # Your code here
    pass

3 test cases · run your code against them in the editor