48 lines
1.4 KiB
Python
48 lines
1.4 KiB
Python
import lzma
|
|
import csv
|
|
import pandas as pd
|
|
from sklearn.linear_model import LinearRegression
|
|
from sklearn.feature_extraction.text import TfidfVectorizer
|
|
from sklearn.pipeline import Pipeline
|
|
|
|
|
|
def readInput(dir):
|
|
X = []
|
|
if 'xz' in dir:
|
|
with lzma.open(dir) as f:
|
|
for line in f:
|
|
text = line.decode('utf-8')
|
|
text = text.split('\t')
|
|
X.append(text)
|
|
else:
|
|
with open(dir, encoding='utf8', errors='ignore') as f:
|
|
for line in f:
|
|
X. append(line.replace('\n',''))
|
|
return X
|
|
|
|
def writeOutput(output, dir):
|
|
with open(dir, 'w', newline='') as f:
|
|
writer = csv.writer(f)
|
|
writer.writerows(output)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
train = pd.DataFrame(readInput('train/train.tsv.xz'),
|
|
columns=['Beginning', 'End', 'Title', 'Source', 'X'])
|
|
train['Y'] = train.apply(lambda x: (float(x.Beginning) + float(x.End))/2, axis=1)
|
|
train = train.drop(columns=['Beginning', 'End', 'Title', 'Source'])
|
|
|
|
model = Pipeline([TfidfVectorizer(), LinearRegression()])
|
|
model.fit(train.X, train.Y)
|
|
|
|
# dev-0
|
|
testX = readInput('dev-0/in.tsv')
|
|
writeOutput(model.predict(testX), 'dev-0/out.tsv')
|
|
|
|
# dev-1
|
|
testX = readInput('dev-1/in.tsv')
|
|
writeOutput(model.predict(testX), 'dev-1/out.tsv')
|
|
|
|
# test-A
|
|
testX = readInput('test-A/in.tsv')
|
|
writeOutput(model.predict(testX), 'test-A/out.tsv') |