Parallel implementation of data subsetting

parallel_subset.py

import numpy as np
from joblib import Parallel, delayed

def subset(file):
    dataOut = []
    data = open(file, 'r')
    data.readline()
    for line in data:
        splitLine = line.split('\t')
        if splitLine[3] == '57':
            dataOut.append(splitLine)
    return dataOut

def stack(list_of_data, hold_data):
    for i in xrange(len(list_of_data)):
        current = np.array(data[i])
        hold = np.vstack((hold_data, current))
    return hold

if __name__ == "__main__":
    filepath = ['testData1.txt', 'testData2.txt']

    hold  = []
    temp = open(filepath[0], 'r')
    hold.append(temp.readline().split('\t'))
    data = Parallel(n_jobs=-1)(delayed(subset)(x) for x in filepath)
    finalData = stack(data, hold)