yoeo
commented
4 years ago Hello @malahmadi1
Yes.
Now you have guess.probabilities(source_code) that gives you the probability for each language.
The first probability can be used as a confidence.
Example:
from pprint import pprint
from guesslang import Guess
guess = Guess()
source_code = """
% Quick sort
-module (recursion).
-export ([qsort/1]).
qsort([]) -> [];
qsort([Pivot|T]) ->
qsort([X || X <- T, X < Pivot])
++ [Pivot] ++
qsort([X || X <- T, X >= Pivot]).
"""
probabilities = guess.probabilities(source_code)
pprint(probabilities)
# Prints the following list:
[('Erlang', 0.7835302948951721),
('Markdown', 0.04323587566614151),
('Haskell', 0.03866693750023842),
('R', 0.027196552604436874),
('Shell', 0.018274815753102303),
('TeX', 0.01470933761447668),
('Matlab', 0.011461317539215088),
('JavaScript', 0.011054938659071922),
('Scala', 0.00946285855025053),
('Perl', 0.004884811118245125),
('C++', 0.0037576162721961737),
('HTML', 0.003617567475885153),
('Rust', 0.0034391707740724087),
('Swift', 0.003167798975482583),
('Ruby', 0.0029752985574305058),
('C', 0.0025446831714361906),
('Objective-C', 0.0023764485958963633),
('Python', 0.0020462924148887396),
('CoffeeScript', 0.001939206849783659),
('Java', 0.0018487452762201428),
('Lua', 0.00134648394305259),
('Jupyter Notebook', 0.0012952813412994146),
('C#', 0.0011168558849021792),
('Go', 0.0011112005449831486),
('Batchfile', 0.00104153947904706),
('TypeScript', 0.0009122725459747016),
('PHP', 0.0008308574906550348),
('SQL', 0.000818912812974304),
('PowerShell', 0.0007574326591566205),
('CSS', 0.0005784708191640675)]
Thank you for your feedback.
Can I obtain a confidence score for each guess?
Thanks, Mohammad