109 lines
3.8 KiB
Python
109 lines
3.8 KiB
Python
import pandas as pd
|
|
import csv
|
|
import sys
|
|
import regex as re
|
|
from collections import Counter, defaultdict
|
|
from nltk import trigrams, word_tokenize
|
|
|
|
def clean_text(text):
|
|
text = text.lower().replace('-\\n', '').replace('\\n', ' ')
|
|
text = re.sub(r'\p{P}', '', text)
|
|
|
|
return text
|
|
|
|
class Model():
|
|
def __init__(self, alpha):
|
|
self.alpha = alpha
|
|
self.probs = defaultdict(lambda: defaultdict(lambda: 0))
|
|
self.vocab = set()
|
|
|
|
def train(self, data):
|
|
for index, row in data.iterrows():
|
|
text = clean_text(str(row['text']))
|
|
words = word_tokenize(text)
|
|
for w1, w2, w3 in trigrams(words, pad_right=True, pad_left=True):
|
|
if w1 and w2 and w3:
|
|
self.vocab.add(w1)
|
|
self.vocab.add(w2)
|
|
self.vocab.add(w3)
|
|
self.probs[(w1, w3)][w2] += 1
|
|
# limit number of data rows used for training
|
|
if index > 10000:
|
|
break
|
|
|
|
for w1_w3 in self.probs:
|
|
total_count = float(sum(self.probs[w1_w3].values()))
|
|
denominator = total_count + self.alpha * len(self.vocab)
|
|
for w2 in self.probs[w1_w3]:
|
|
nominator = self.probs[w1_w3][w2] + self.alpha
|
|
self.probs[w1_w3][w2] = nominator / denominator
|
|
|
|
def predict(self, w1, w3):
|
|
raw_prediction = dict(self.probs[w1, w3])
|
|
prediction = dict(Counter(raw_prediction).most_common(6))
|
|
|
|
total_prob = 0.0
|
|
str_prediction = ''
|
|
|
|
for word, prob in prediction.items():
|
|
total_prob += prob
|
|
str_prediction += f'{word}:{prob} '
|
|
|
|
remaining_prob = 1 - total_prob
|
|
|
|
str_prediction += f':{remaining_prob}'
|
|
|
|
return str_prediction
|
|
|
|
|
|
# check arguments
|
|
if len(sys.argv) != 2:
|
|
print('Wrong number of arguments. Expected 1 - alpha smoothing parameter.')
|
|
quit()
|
|
else:
|
|
alpha = sys.argv[1]
|
|
|
|
# load training data
|
|
train_data = pd.read_csv('train/in.tsv.xz', sep='\t', error_bad_lines=False, warn_bad_lines=False, header=None, quoting=csv.QUOTE_NONE)
|
|
train_labels = pd.read_csv('train/expected.tsv', sep='\t', error_bad_lines=False, warn_bad_lines=False, header=None, quoting=csv.QUOTE_NONE)
|
|
|
|
train_data = train_data[[6, 7]]
|
|
train_data = pd.concat([train_data, train_labels], axis=1)
|
|
|
|
train_data['text'] = train_data[6] + train_data[0] + train_data[7]
|
|
train_data = train_data[['text']]
|
|
|
|
# init model with given aplha
|
|
model = Model(alpha)
|
|
|
|
# train model probs
|
|
model.train(train_data)
|
|
|
|
# make predictions
|
|
dev_data = pd.read_csv('dev-0/in.tsv.xz', sep='\t', error_bad_lines=False, warn_bad_lines=False, header=None, quoting=csv.QUOTE_NONE)
|
|
test_data = pd.read_csv('test-A/in.tsv.xz', sep='\t', error_bad_lines=False, warn_bad_lines=False, header=None, quoting=csv.QUOTE_NONE)
|
|
|
|
with open('dev-0/out.tsv', 'w') as file:
|
|
for index, row in dev_data.iterrows():
|
|
left_text = clean_text(str(row[6]))
|
|
right_text = clean_text(str(row[7]))
|
|
left_words = word_tokenize(left_text)
|
|
right_words = word_tokenize(right_text)
|
|
if len(left_words) < 2 or len(right_words) < 2:
|
|
prediction = ':1.0'
|
|
else:
|
|
prediction = model.predict(left_words[len(left_words) - 1], right_words[0])
|
|
file.write(prediction + '\n')
|
|
|
|
with open('test-A/out.tsv', 'w') as file:
|
|
for index, row in test_data.iterrows():
|
|
left_text = clean_text(str(row[6]))
|
|
right_text = clean_text(str(row[7]))
|
|
left_words = word_tokenize(left_text)
|
|
right_words = word_tokenize(right_text)
|
|
if len(left_words) < 2 or len(right_words) < 2:
|
|
prediction = ':1.0'
|
|
else:
|
|
prediction = model.predict(left_words[len(left_words) - 1], right_words[0])
|
|
file.write(prediction + '\n')
|