MediumNLPnlp-370

Weighted Average Word Embeddings

Problem

Weighted pooling computes sum_i w_i v_i / sum_i w_i.

Given the word vectors and their weights, return the weighted mean vector.

Hint

Weights need not sum to one; normalize explicitly.

case 1

in: [[[1,0],[0,1]],[3,1]]

expected: [0.75,0.25]

case 2

in: [[[2,2]],[1]]

expected: [2,2]

case 3

in: [[[1,1],[3,3]],[1,3]]

expected: [2.5,2.5]

Starter codePython

def weighted_average_word_embeddings(vectors, weights):
    # Your code here
    pass

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