import sklearn.metrics
from sklearn.cluster import KMeans
from sklearn.feature_extraction.text import TfidfVectorizer

import numpy as np

with open("dev-0/in.tsv") as myFile:
    cnt = myFile.readlines()
    vectorizer = TfidfVectorizer(ngram_range = (1,2), use_idf = False)
    vectorizer = TfidfVectorizer()
    document_vectors = vectorizer.fit_transform(cnt)
    kmeans = KMeans(n_clusters = 77, random_state = 0).fit(document_vectors)
    pred_myFile = kmeans.fit_predict(document_vectors)
    tmp = kmeans.labels_
    tmp = np.array2string(tmp, precision = 2, separator = '\n',suppress_small = True)
    myFile = open("dev-0/out.tsv", "a")
    myFile.write(tmp)