In [1]:

    

import os
import pandas as pd
import gzip
import load


    

In [5]:

    

def combine_features(house, features):
    fname = '../data/kasteren/2010/datasets/house{house}/{feature}.csv.gz'

    files = map(lambda f: fname.format(house=house, feature=f), features)
    data_arr = map(lambda loc: load.data(loc), files)
    arr = []
    for i, d in enumerate(data_arr):
        if i != len(data_arr) - 1:
            del d['activity']
        arr.append(d)

    df = pd.concat(arr, axis = 1)

    print len(df.columns)
    f = '../data/kasteren/2010/datasets/house{house}/{feature}.csv.gz'.format(house=house, feature=''.join(features))

    df.to_csv(gzip.open(f, 'w'))


    

In [6]:

    

combine_features('A',['last', 'change'])


    

    




29

In [7]:

    

combine_features('B',['last', 'change'])
combine_features('C',['last', 'change'])


    

    




57
47

In [ ]: