from sklearn.feature_extraction.text import TfidfVectorizer from sklearn.linear_model import LinearRegression import string import csv dev = [] with open("out.tsv", 'r', encoding="utf-8") as out: dev = [line.strip() for line in out] i = -1 with open("fix.tsv", 'w', encoding="utf-8") as fix: for score in dev: score = int(float(score)) fix.write(str(score) + ".5\n")