EasyNLPnlp-356

Most Similar Token by Cosine

Problem

Given a target vector and a dictionary of candidate name to vector, return the candidate name with the highest cosine similarity. Return None when the dictionary is empty.

Given the target vector and candidates, return the best name.

Hint

Score each candidate and track the best name.

Test cases

case 1

in: [[1,0],{"x":[1,0],"y":[0,1]}]

expected: "x"

case 2

in: [[1,1],{"a":[1,0],"b":[1,1]}]

expected: "b"

case 3

in: [[1,0],{}]

expected: null

Starter codePython

def most_similar_token_by_cosine(target, candidates):
    # Your code here
    pass

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