Python - Looping Issue

Question

I am using a for loop to update a Numpy Array but it doesn't seem to be working. What am I doing wrong?

My code

import pandas as pd
import pandas_datareader as pdr
import datetime as dt
from dateutil.relativedelta import relativedelta
import matplotlib.finance as mf
import numpy as np
import scipy as sp

def pull_price(ticker):

    df = pd.read_csv(ticker+'.csv')
    df1=np.array(df['Adj Close'])
    return df1




tickers=['^DJI','^GSPC','^IXIC','^GDAXI','^HSI','^FCHI','^N225']

dic={}

for i in tickers:
    dic[i]=pull_price(i)


for i in tickers:
    count = len(dic[i])
    for j in range(0,count):
        try:
            dic[i][j] = float(dic[i][j])
        except ValueError:
            dic[i][j] = float(dic[i][j-1])

dic_1= dic
print(dic['^GSPC'][0])

for i in tickers:
    count = len(dic_1[i])
    for j in range(0,count):
        dic_1[i][j] = np.log(dic_1[i][j])

print(dic_1['^GSPC'][0])
print(dic['^GSPC'][0])

My output for the 3 Print commands in sequence

411.410004
6.01959029389
6.01959029389

For the 1st Print command it shows me the correct value. But for the next 2 Print Commands why is it showing the same value. I am only updating dic_1 and not dic. What am I doing wrong?

Answer 1

When you do:

dict1 = dict(a=1)
dict2 = dict1

Both dict1 and dict2 point to the same dictionary. The names are different, but both names refer to the same data structure. That means:

dict2['a'] = 2
print(dict1)
{'a': 2}

If you want dict2 to be a copy of dict1, explicitly make it a copy:

dict2 = dict1.copy()

Ah, and because the elements of your dictionary are themselves also dictionaries, you need to make a deep copy:

import deepcopy
dict2 = deepcopy.copy(dict1)