while statement giving 'float' object has no attribute 'getitem' error using numpy

Question

I am trying to calculate the Average True Range of a data series which has been read and parsed from a .csv file. my code is as follows:

import datetime
import time
import matplotlib.pyplot as plt
import numpy as np

fhand = open('C:\Users\Stuart\Desktop\FX Programming\EURUSD_hour.csv', 'r')



for line in fhand:
    line = line.split(',') 
    fxpair, _date, _time, _open, _high, _low, _close = line[0], line[1], line[2], line[3], float(line[4]), float(line[5]), float(line[6])
    date_time = datetime.datetime.strptime('{} {}'.format(_date.partition(' ')[0], _time),'%Y%m%d %H:%M:%S')



#define Average True Range function    
def TR(d,c,h,l,o,yc):
    x = h-l
    y = abs(h-yc)
    z = abs(l-yc)

    print x
    print y
    print z

    if y <= x >= z:
        TR = x
    elif x <= y >= z:
        TR = y
    elif x <= z >= y:
        TR = z

    print d, TR
    return d, TR

x = 1

TRDates = []
TrueRanges = []

while x < len(_date):
    TRDate, TrueRange = TR(_date[x],_close[x],_high[x],_low[x],_open[x],_close[x-1])
    TRDates.append(TRDate)
    TrueRanges.append(TrueRange)
    x+=1   

def ExpMovingAverage(values, window):
    weights = np.exp(np.linspace(-1., 0., window))
    weights /= weights.sum()
    a =  np.convolve(values, weights, mode='full')[:len(values)]
    a[:window] = a[window]
    return a


print len(TrueRanges)
ATR = ExpMovingAverage(TrueRanges,14)

print ATR

However I am getting the following error message:

     46 while x < len(line):
---> 47     TRDate, TrueRange = TR(_date[x],_close[x],_high[x],_low[x],_open[x],_close[x-1])
     48     TRDates.append(TRDate)
     49     TrueRanges.append(TrueRange)

TypeError: 'float' object has no attribute '__getitem__'

So there is obviously something wrong with the while statement, and it isnt iterating over the _date like I want it to. I have also tried using date_time instead of _date but then I get:

---> 46 while x < len(date_time):
     47     TRDate, TrueRange = TR(date_time[x],_close[x],_high[x],_low[x],_open[x],_close[x-1])
     48     TRDates.append(TRDate)

TypeError: object of type 'datetime.datetime' has no len()

Can someone help me to get this working?

I have a very similar code which I have pasted below that actually works, which uses a different text file for data and unpacks it a different way. Can someone help me reconcile the two so that the top code works?

import numpy as np


sampleData = open('C:\\Users\\Stuart\\Desktop\\FX Programming\\sampleData.txt','r').read()
splitData = sampleData.split('\n')

date,closep,highp,lowp,openp,volume = np.loadtxt(splitData, delimiter=',',unpack=True)

def TR(d,c,h,l,o,yc):
    x = h-l
    y = abs(h-yc)
    z = abs(l-yc)

    print x
    print y
    print z

    if y <= x >= z:
        TR = x
    elif x <= y >= z:
        TR = y
    elif x <= z >= y:
        TR = z

    print d, TR
    return d, TR

x = 1

TRDates = []
TrueRanges = []

while x < len(date):
    TRDate, TrueRange = TR(date[x],closep[x],highp[x],lowp[x],openp[x],closep[x-1])
    TRDates.append(TRDate)
    TrueRanges.append(TrueRange)
    x+=1



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

def ExpMovingAverage(values, window):
    weights = np.exp(np.linspace(-1., 0., window))
    weights /= weights.sum()
    a =  np.convolve(values, weights, mode='full')[:len(values)]
    a[:window] = a[window]
    return a


print len(TrueRanges)
ATR = ExpMovingAverage(TrueRanges,14)

print ATR

All that error message means is that one of the inputs to TR, which you attempt to index with [x] is a number, not an array or list. Before jumping to using TR, make sure that the for line in fhand: block is producing the desired values (lists of numbers?). Off hand it looks like those variables might be strings. — hpaulj
– hpaulj, Commented Jul 10, 2014 at 16:19

Bi Rico · Accepted Answer · 2014-07-10 16:07:31Z

1

In the code that works, high low and close are all arrays so you can index them by date. For example (I used a list instead of an array, but it's similar):

hi = [10, 11, 12]
print hi[0]
# 10

In your code you're looping over your file, converting these values to float and then discarding them. Here is a simple example to demonstrate:

for value in ['10', '11', '12']
    x = float(value)
print x
# 12

Notice that each assignment replaces the existing x and sets x to a new float. When you finally print x you get the last thing it was assigned to. You need to rewrite the parsing of your text file to save all the values. Maybe something like:

x = []
for value in ['10', '11', '12']
    x.append(float(value))

answered Jul 10, 2014 at 16:07

Bi Rico

25.9k3 gold badges57 silver badges75 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

s666 Over a year ago

Thanks I have changed my code to append each individual "open, high, low, close, date" to a separate array o = [], h = [], l = [] etc. Now I manage to get past that line, but encounter an "index out of bounds" error on the line ---> 61 a[:window] = a[window]. But i guess that is a question for another page...thanks very much for your help!

Collectives™ on Stack Overflow

while statement giving 'float' object has no attribute 'getitem' error using numpy

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related