57 lines
2.0 KiB
Python
57 lines
2.0 KiB
Python
import string
|
|
from gensim.models.doc2vec import Doc2Vec, TaggedDocument
|
|
from sklearn.cluster import KMeans
|
|
from sklearn.naive_bayes import MultinomialNB
|
|
from sklearn.preprocessing import MinMaxScaler, normalize
|
|
|
|
scaler = MinMaxScaler()
|
|
|
|
stopwords = []
|
|
# stopwords source - https://github.com/bieli/stopwords/blob/master/polish.stopwords.txt
|
|
with open('stopwords') as f:
|
|
stopwords = [line.rstrip() for line in f]
|
|
|
|
classifier = MultinomialNB()
|
|
|
|
def train():
|
|
with open('train/train.tsv') as f:
|
|
docs = [line.rstrip() for line in f]
|
|
docs_preprocessed = []
|
|
y = []
|
|
for doc in docs:
|
|
t = doc.split('\t')
|
|
y.append(t[0])
|
|
doc = t[1]
|
|
doc = doc.lower().split(' ')
|
|
doc = [''.join(char for char in word if char not in string.punctuation) for word in doc]
|
|
doc = list(filter(lambda word: (word not in stopwords) and (word != ''), doc))
|
|
docs_preprocessed.append(doc)
|
|
y = [int(numeric_string) for numeric_string in y]
|
|
tagged_documents = [TaggedDocument(doc, [i]) for i, doc in enumerate(docs_preprocessed)]
|
|
global d2v_model
|
|
d2v_model = Doc2Vec(tagged_documents, epochs=300, dm=0)
|
|
X = d2v_model.dv.vectors
|
|
X = scaler.fit_transform(X)
|
|
classifier.fit(X, y)
|
|
|
|
def classify(path):
|
|
with open(path + 'in.tsv') as f:
|
|
docs = [line.rstrip() for line in f]
|
|
docs_preprocessed = []
|
|
for doc in docs:
|
|
doc = doc.lower().split(' ')
|
|
doc = [''.join(char for char in word if char not in string.punctuation) for word in doc]
|
|
doc = list(filter(lambda word: (word not in stopwords) and (word != ''), doc))
|
|
docs_preprocessed.append(doc)
|
|
test_vectors = []
|
|
for doc in docs_preprocessed:
|
|
test_vectors.append(d2v_model.infer_vector(doc))
|
|
results = classifier.predict(test_vectors)
|
|
with open(path + 'out.tsv', 'w') as file:
|
|
for result in results:
|
|
file.write("%i\n" % result)
|
|
|
|
|
|
train()
|
|
classify('dev-0/')
|
|
# classify('test-A/', n_clusters=10) |