# Importi
import csv
import time
from time import mktime
import pandas as pd
from datetime import datetime
import numpy as np
import matplotlib.pyplot as plt
import matplotlib as mpl
path='/home/rok/muzej/analiza/obrazi1.csv'
path_new='/home/rok/muzej/analiza/obrazi_new.csv'
with open(path_new, 'w') as f_new:
writer = csv.writer(f_new, delimiter=',')
writer.writerow(['Datum', 'Skupaj'])
with open(path, 'rb') as f:
reader = csv.DictReader(f)
for row in reader:
if row['Datum'] != '' and row['Datum'] != ' ' and row['SKUPAJ'] != '':
writer.writerow([row['Datum'], row['SKUPAJ']])
# V path_new je sedaj pot do datoteke za analizo
with open(path_new, 'rb') as f:
datumi_obisk = []
reader = csv.DictReader(f)
for row in reader:
datumi_obisk.append(row)
# Primer podatkov
print datumi_obisk[:3]
df = pd.DataFrame(datumi_obisk)
df.head()
# Metoda za pridobitev dneva
def get_day(d):
return datetime.fromtimestamp(mktime(d)).weekday()
# Dnevi so označeni s številkami od 0 (Ponedeljek) do 6 (Nedelja)
dnevi_obisk = []
for row in datumi_obisk:
day = get_day(time.strptime(row['Datum'], "%m/%d/%Y"))
dnevi_obisk.append({'Dan': day, 'Skupaj': int(row['Skupaj'])})
print dnevi_obisk[:3]
dnevi_df = pd.DataFrame(dnevi_obisk)
dnevi_df.head()
grps = dnevi_df.groupby(by=['Dan'])['Skupaj']
desc = grps.agg(['sum','count','mean','median','min','max','std','var'])
print desc