Return the cumulative sum of array elements over given axis treating NaNs as zero in Python

To return the cumulative sum of array elements over a given axis treating NaNs as zero, use the nancumsum() method. The cumulative sum does not change when NaNs are encountered and leading NaNs are replaced by zeros.

The method returns a new array with cumulative sums computed along the specified axis. Zeros are returned for slices that are all-NaN or empty. Cumulative sum works progressively: 5, 5+10, 5+10+15, 5+10+15+20.

Syntax

numpy.nancumsum(a, axis=None, dtype=None, out=None)

Parameters

• a − Input array
• axis − Axis along which the cumulative sum is computed. Default (None) computes over flattened array
• dtype − Type of returned array and accumulator. Defaults to input array's dtype
• out − Alternative output array to place the result

Example

Let's create a 2D array with NaN values and compute cumulative sum along different axes ?

import numpy as np

# Creating a numpy array with NaN values
arr = np.array([[10, 20, 30], [40, np.nan, 60]])

print("Our Array...")
print(arr)

print("\nDimensions:", arr.ndim)
print("Datatype:", arr.dtype)

# Cumulative sum along axis=1 (rows)
print("\nCumulative Sum along axis=1:")
print(np.nancumsum(arr, axis=1))

Our Array...
[[10. 20. 30.]
 [40. nan 60.]]

Dimensions: 2
Datatype: float64

Cumulative Sum along axis=1:
[[ 10.  30.  60.]
 [ 40.  40. 100.]]

Different Axis Examples

Let's see cumulative sum along different axes and with flattened array ?

import numpy as np

arr = np.array([[5, np.nan, 15], [20, 25, np.nan]])
print("Array:")
print(arr)

# Along axis=0 (columns)
print("\nCumulative sum along axis=0:")
print(np.nancumsum(arr, axis=0))

# Along axis=1 (rows) 
print("\nCumulative sum along axis=1:")
print(np.nancumsum(arr, axis=1))

# Flattened array (axis=None)
print("\nCumulative sum of flattened array:")
print(np.nancumsum(arr))

Array:
[[ 5. nan 15.]
 [20. 25. nan]]

Cumulative sum along axis=0:
[[ 5. nan 15.]
 [25. 25. 15.]]

Cumulative sum along axis=1:
[[ 5.  5. 20.]
 [20. 45. 45.]]

Cumulative sum of flattened array:
[ 5.  5. 20. 40. 65. 65.]

Comparison with Regular cumsum()

import numpy as np

arr = np.array([1, 2, np.nan, 4])

print("Regular cumsum():")
print(np.cumsum(arr))

print("\nnancumsum():")
print(np.nancumsum(arr))

Regular cumsum():
[ 1.  3. nan nan]

nancumsum():
[1. 3. 3. 7.]

Conclusion

Use np.nancumsum() to compute cumulative sums while treating NaN values as zero. This is especially useful in data analysis where missing values should not break the cumulative calculation.