munging

import re
import pandas as pd
import numpy as np
df = pd.read_csv('/Users/dk1306/tutorial/tutorial/items.csv')
df.keys()
df.shape
np.sum(df.isnull())          #missing values each column
np.sum(df.isnull(), axis=1)  #missing values each row
df.isnull().any(axis=1)
df.loc[df.isnull().any(axis=1),:]
df = df.dropna(axis=0, how='any')
df.shape
df.describe()
df.groupby('category').size()



##################################################

import datetime
def date_days(DATE):
    DATE = DATE[8:]
    return (datetime.date.today() -datetime.datetime.strptime(DATE, '%B %d, %Y').date()).days

df['days'] = map(date_days,df['date'])

##################################################

df['people'] = map(lambda a: int(a.replace(',', '')),df['people'])


##################################################

def current_number(current):
        #print current,float(current.replace(',', ''))
        if (float(current.replace(',', ''))<10):
            return int(float(current.replace(',', ''))*1000)
        return int(current.replace(',', ''))
    
df['current']=map(current_number,df['current'])

##################################################





def shares_number(shares):
    
    shares = shares[2:]
    if (shares[-1] == 'K'):
        return int(float(shares[:-1])*1000)
    else:
        return int(shares )
    
df['shares'] = map(shares_number,df['shares'])

##################################################

def target_selection(target):
    target = target[5:-5]
    return target 

def target_number(target):
    if(target[0] =='\x82'):
        target = target.replace('\x82\xac', '')
    if(target[0] == '\xa3'):
        target = target.replace('\xa3', '')
    if (target[-1] == 'k'):
        return int(float(target[:-1])*1000)
    if (target[-1] == 'M'):
        return int(float(target[:-1])*1000000)
    if (float(target.replace(',', ''))<10):
        return int(float(target.replace(',', ''))*1000)
    else :
        return int(target.replace(',', ''))
    
df['target']=map(target_number,map(target_selection,df['target']))

##################################################

df['percent_complt'] = abs((df['target']-df['current'])/df['target'] )*100

##################################################

df['avg_contrb'] = df['current']/df['people']

##################################################

df.describe()

data = pd.concat([df['category'],df['target'], df['current'], df['percent_complt'], df['people'],df['avg_contrb'], df['shares'],df['days']], axis=1,
          keys=['category','target','current','percent_complt', 'people','avg_contrb','shares','days'])


data.to_csv('gofundme.csv')
df.describe()
data.head(5)