local variable referenced before assignment | matplotlib

Question

I want to visualize the central limit theorem for an exemplary PDF. The following code works as I wish:

import sympy as sp
import numpy as np
import matplotlib.pyplot as plt

x, y = sp.symbols('x y')

dens = sp.exp(-x) #PDF

def sum(dens,n):
    """
    Parameters
    ----------
    dens : probabilty density function
    n : amount of iteration

    Returns
    -------
    pdf after n iterations

    """
    dens2 = dens.subs(x, y-x) #PDF for one further summed random variable
    
    i=1
    while i<=n:
        int = sp.integrate(dens*dens2, (x,0,y))
        int = int.subs(y,x)
        
        dens = int
        i += 1
        
    return int

#plot for n th iteration
n1 = 1
n2 = 20
n3=50
n4 = 100


X1 = np.linspace(0,200,num=1000)
Y1 = sp.lambdify(x,sum(dens,n1),'numpy')
plt.plot(X1, Y1(X1),label="n=1")


X2 = np.linspace(0,200,num=1000)
Y2 = sp.lambdify(x,sum(dens,n2),'numpy')
plt.plot(X2, Y2(X2),label="n=20")

X3 = np.linspace(0,200,num=1000)
Y3 = sp.lambdify(x,sum(dens,n3),'numpy')
plt.plot(X3, Y3(X3),label="n=50")

X4 = np.linspace(0,200,num=1000)
Y4 = sp.lambdify(x,sum(dens,n4),'numpy')
plt.plot(X4, Y4(X4),label="n=100")


plt.legend()
plt.show()

Now I'd like to do the plot for all the n possibilities (later I want to try to animate it, but at first I need to understand how to do this loop). Thus I want to do the plot using a loop instead of creating the plots separately as above. But this gives me the error

Traceback (most recent call last):

File "C:\Users\user\Desktop\ZGS.py", line 71, in Y = sp.lambdify(x,sum(dens,k),'numpy')

File "C:\Users\user\Desktop\ZGS.py", line 32, in sum return int

UnboundLocalError: local variable 'int' referenced before assignment

I tried some things such as global int but this creates problems within sympy. Why can I use different variables for n when plotting separately but get this error when assigning n using a loop?

n=100
for k in range(n):
    X = np.linspace(0,200,num=1000)
    Y = sp.lambdify(x,sum(dens,k),'numpy')
    plt.plot(X, Y(X))
    
    plt.show()

How can this problem be solved?

`while i<=n:` - is it possible the function was called with `i>n`? [Catch the error](https://docs.python.org/3/tutorial/errors.html#handling-exceptions) and inspect/print relevant data in the except suite. If you are using an IDE **now** is a good time to learn its debugging features Or the built-in [Python debugger](https://docs.python.org/3/library/pdb.html). [What is a debugger and how can it help me diagnose problems?](https://stackoverflow.com/questions/25385173/what-is-a-debugger-and-how-can-it-help-me-diagnose-problems) — wwii, Dec 13 '20 at 16:18
@JohanC It's not a keyword. If it were, this mistake wouldn't happen. — mkrieger1, Dec 13 '20 at 16:29
@mkrieger1 Well, strictly speaking `int` would be a 'built-in class' and `sum` a 'built-in utility function'. Using a variable with the same name usually isn't a good idea. — JohanC, Dec 13 '20 at 16:45

wwii · Answer 1 · 2020-12-13T16:29:19.090

for k in range(n):
   ...
   Y = sp.lambdify(x,sum(dens,k),'numpy')
   ...

On the first iteration k is zero.

>>> for k in range(3):
...     print(k)
... 
0
1
2

When dens is called with k == 0 - while i<=n is False and nothing in that while loop is processed. When return int is processed int does not exist.

range takes an optional start argument which would alleviate your error:

>>> for k in range(1,3+1):
...     print(k)
... 
1
2
3
>>>

local variable referenced before assignment | matplotlib

1 Answers1