"AttributeError: 'NoneType' 'write'" & "RuntimeWarning: overflow double_scalars" - received during a while loop

Question

I'm using a while loop to obtain a backwards decay curve for a radioactive element.

import numpy as np
import csv

lmbd=np.log(2)/12.43
year=list(range(-48050,2012))

f=0
decay=[] 
decay.append(0.45)  # concentration of sample during year of sampling

while f<len(year): # Creating the backwards decay curve sample
    p=decay[f]
    x=p/np.exp(-lmbd)
    decay.append(x)
    f=f+1

print(decay)

When I run the above code I get the following error:

Traceback (most recent call last): File "", line 13, in

x=p/np.exp(-lmbd)  

File "C:\Python32\lib\idlelib\PyShell.py", line 59, in idle_showwarning file.write(warnings.formatwarning(message, category, filename,

AttributeError: 'NoneType' object has no attribute 'write'

However, if I substitute len(year) with, for example, 50 then the loop runs smoothly. If I substitute len(year) with 50062 (the value of len(year)) I still get the error.

1. Could the error occurring due to the length of the iterations? (though this surely mustn't be the case)
2. Why is it working for 50 but not 50062?
3. Can it be that the numbers just get too big as I'm creating exponential growth over 50000 years? (I wanted to keep that many years as I use the same script for other elements which have a smaller lmbd value so values don't get so large)
4. Any possibilities for fixing?

I'm using Python 3.2

Answer 1

Thanks to PaF, User, and eryksun for providing the answers (in the comments) :

1. I was getting the wrong warning (in IDLE) in regards to the script.

   • If I ran the script in cmd (as per User suggestions) I got the RuntimeWarning: overflow encountered in double_scalars x=p/np.exp(-lmbd) error.

   • Also if I added import sys, idlelib.PyShell; idlelib.PyShell.warning_stream = sys.stderr to the script (as per eruksun suggestion) I also get the correct error in IDLE (which I was using).

   • The script can now run to completion but the decay list has a whole lot of inf values at the end which is dealt with in the next point.

2. To deal with the overflow problem:

   • I tried replacing the while statement with while f<len(year) and decay[-1] != np.inf: (as per PaF suggestion). The loop stops once the last value in the decay list reaches inf; however, it attaches a single inf value to the end of the list. Additionally, the overflow error remains.

   • I instead added an if statement within the loop to exit once inf is reached but before appending to the list (however, the overflow error still comes up but produces a list without any inf values in it)

The updated script, which works, is:

import numpy as np
import csv
import sys, idlelib.PyShell; idlelib.PyShell.warning_stream = sys.stderr

lmbd=np.log(2)/12.43
year=list(range(-48050,2012))

f=0
decay=[] 
decay.append(0.45)# concentration of sample during year of sampling

while f<len(year):  # Creating the backwards decay equation for each sample
    p=decay[f]
    x=p/np.exp(-lmbd)
    if x==np.inf:
        break
    decay.append(x)
    f=f+1

print(decay)
print("end")

While this still gives the overflow error in both IDLE and CMD, I think that is OK for my purposes as I'm not worried about the extremely large numbers and the script runs to completion.