Solved – runtimewarning: invalid value encountered in double_scalars

In this article, you will learn how to solve runtimewarning: invalid value encountered in double_scalars.

Let’s look at a code example that produces the same error.

from numpy import cosh
print("3. ",  0.3+0.2-0.5)  
print("4. ",  0.1+0.2-0.3)  

x = 30
print("3. ", (1-cosh(x))/(1+cosh(x)))
x = 900
print("7. ", (1-cosh(x))/(1+cosh(x))) # this produces overflow error

Output of above code:

3.  0.0
4.  5.551115123125783e-17
3.  -0.9999999999996257
<string>:8: RuntimeWarning: overflow encountered in cosh
<string>:8: RuntimeWarning: invalid value encountered in double_scalars
7.  nan
> 

The code snippet above is actually running into Divide by Zero Error. When the value or result is larger than the declared operation or data type in the program then this throws an overflow error indicating the value is exceeding the given or declared limit value.

how to solve runtimewarning: invalid value encountered in double_scalars.

you can use scipy.special.logsumexp:

In [52]: from scipy.special import logsumexp

In [53]: res = np.exp(logsumexp(-3*d) - logsumexp(-3*e))

In [54]: res
Out[54]: 1.1050349147204485e-116

Alternatively, you can  disable run time warnings (not recommended) like below:

import warnings

def f(x, y):
    with warnings.catch_warnings():
        warnings.simplefilter('ignore')        
        return ((x**(2))*y)/((x**(4)) + (y**(4)))