MediumNLPnlp-401

Smooth Inverse Document Frequency Value

Problem

A smoothed IDF is ln((N + 1) / (df + 1)) + 1, which stays positive for every term.

Given the number of documents N and the document frequency df, return the smooth IDF.

Hint

The +1 shift keeps rare and common terms comparable.

case 1

in: [100,10]

expected: 3.217225244042889

case 2

in: [5,5]

expected: 1

case 3

in: [1000,1]

expected: 7.215607598755275

Starter codePython

def smooth_inverse_document_frequency_value(n_docs, df):
    # Your code here
    pass

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