NumPy RuntimeWarning: divide by zero encountered in log10

# NumPy RuntimeWarning: divide by zero encountered in log10

The NumPy "RuntimeWarning: divide by zero encountered in log10" is shown when you pass an array that contains zeros to the numpy.log10() method.

To resolve the issue, use the numpy.seterr() method to disable the warnings or use a context manager.

runtimewarning divide by zero encountered in log10

Here is an example of how the warning is shown.

main.py

Copied!
import numpy as np

arr = np.array([4, 12, 0, 16, 160, 320])

# /home/borislav/Desktop/bobbyhadz_python/main.py:14:
# RuntimeWarning: divide by zero encountered in log10
print(np.log10(arr))

The array contains a zero value which is the cause of the warning.

The numpy.log10() method returns the base 10 logarithm of the supplied array, element-wise.

One way to resolve the issue is to use a context manager to set the error state for the division to ignore.

main.py

Copied!
import numpy as np

arr = np.array([4, 12, 0, 16, 160, 320])

with np.errstate(divide='ignore'):
    print(np.log10(arr))

using context manager to ignore division errors

The code for this article is available on GitHub

The numpy.errstate context manager is used for floating-point error handling.

The context manager uses the numpy.seterr() method under the hood.

We set the divide keyword argument to ignore, so NumPy is instructed to ignore division errors in the indented code block.

The divide keyword argument is used to determine how division by zero is treated.

Once we exit the context manager, the error-handling behavior is reverted to the default.

main.py

Copied!
import numpy as np

arr = np.array([4, 12, 0, 16, 160, 320])

with np.errstate(divide='ignore'):
    # 👇️ No error here
    print(np.log10(arr))

# 👇️ This causes the error
print(np.log10(arr))

As shown in the code sample, once we exit the context, the error-handling behavior is reset.

# Using the `seterr` method to disable the warning

You can also use the numpy.seterr() method to manually disable the warning.

main.py

Copied!
import numpy as np

arr = np.array([4, 12, 0, 16, 160, 320])

np.seterr(divide='ignore')

print(np.log10(arr))

np.seterr(divide='warn')

manually disable division by zero errors

The code for this article is available on GitHub

The numpy.seterr() method determines how floating-point errors are handled.

The divide argument is used to set the behavior when dividing by zero.

When the argument is set to ignore, no action is taken when a division by zero exception occurs.

We set the divide keyword argument to warn in the second call to the np.seterr() method.

main.py

Copied!
np.seterr(divide='warn')

This resets the error-handling behavior.

When the keyword argument is set to warn, a RuntimeWarning is printed when a division by zero exception occurs.

If you use the np.log10() method with an array containing zeros after the error handling behavior is reset, you'd get the warning.

main.py

Copied!
import numpy as np

arr = np.array([4, 12, 0, 16, 160, 320])

np.seterr(divide='ignore')

# 👇️ This doesn't raise the warning
print(np.log10(arr))

np.seterr(divide='warn')

# 👇️ This causes the warning
print(np.log10(arr))

The second call to numpy.log10() shows the warning.

# Using the `where` keyword argument to resolve the issue

You can also set the where keyword argument in the call to the numpy.log10() method to resolve the issue.

main.py

Copied!
import numpy as np

arr = np.array([4, 12, 0, 16, 160, 320])

result = np.log10(arr, where=arr > 0)

print(result)

The code for this article is available on GitHub

The where keyword argument checks each array element for the specified condition.

The numpy.log10() method is only applied to the non-zero values in the array.

I've also written an article on how to convert a NumPy array to 0 or 1 based on a threshold.

# Additional Resources

You can learn more about the related topics by checking out the following tutorials:

I wrote a book in which I share everything I know about how to become a better, more efficient programmer.

You can use the search field on my Home Page to filter through all of my articles.

NumPy RuntimeWarning: divide by zero encountered in log10