np.sum() and axis parameter in NumPy - Step-by-Step Execution

Concept Flow - np.sum() and axis parameter

Start with array

↓

Choose axis parameter

↓

Sum elements along axis

↓

Return summed array or scalar

↓

End

np.sum() adds numbers in an array. The axis parameter decides which direction to add: rows, columns, or all.

Execution Sample

NumPy

import numpy as np
arr = np.array([[1, 2], [3, 4]])
sum_all = np.sum(arr)
sum_axis0 = np.sum(arr, axis=0)
sum_axis1 = np.sum(arr, axis=1)

This code sums all elements, then sums by columns (axis=0), then by rows (axis=1).

Execution Table

Step	Operation	Input Array	Axis Parameter	Result
1	Sum all elements	[[1, 2], [3, 4]]	None	10
2	Sum along axis=0 (columns)	[[1, 2], [3, 4]]	0	[4, 6]
3	Sum along axis=1 (rows)	[[1, 2], [3, 4]]	1	[3, 7]

💡 All sums computed; axis parameter controls direction of summation.

Variable Tracker

Variable	Start	After Step 1	After Step 2	After Step 3
arr	[[1, 2], [3, 4]]	[[1, 2], [3, 4]]	[[1, 2], [3, 4]]	[[1, 2], [3, 4]]
sum_all	undefined	10	10	10
sum_axis0	undefined	undefined	[4, 6]	[4, 6]
sum_axis1	undefined	undefined	undefined	[3, 7]

Key Moments - 3 Insights

Why does np.sum(arr) return a single number but np.sum(arr, axis=0) returns an array?

np.sum(arr) adds all numbers in the whole array into one number (see step 1). When axis=0, it sums down each column separately, so the result is an array with one sum per column (step 2).

What does axis=1 mean in np.sum(arr, axis=1)?

Axis=1 means sum across each row. So it adds numbers in each row separately, giving one sum per row (step 3).

What happens if axis is set to a value not in the array's dimensions?

Numpy will raise an error because the axis does not exist. The axis must be 0 or 1 for a 2D array like this.

Visual Quiz - 3 Questions

Look at the execution table, what is the result of np.sum(arr, axis=0)?

A [3, 7]

B [4, 6]

C 10

D [1, 2]

💡 Hint

Check row 2 in the execution table under Result.

At which step does np.sum() add all elements into one number?

A Step 2

B Step 3

C Step 1

D None

💡 Hint

Look at the Operation column for 'Sum all elements'.

If we change axis=1 to axis=0, how does the result change compared to step 3?

A Result sums columns instead of rows

B Result sums rows instead of columns

C Result becomes a single number

D Result stays the same

💡 Hint

Compare results in step 2 and step 3 in the execution table.

Concept Snapshot

np.sum(array, axis=None)
- Sums all elements if axis=None
- axis=0 sums down columns
- axis=1 sums across rows
- Result shape depends on axis
- Useful for quick totals in arrays

Full Transcript

This lesson shows how np.sum() adds numbers in arrays. Without axis, it adds all numbers into one total. With axis=0, it adds numbers down each column separately. With axis=1, it adds numbers across each row separately. The axis parameter controls the direction of summation. We traced a 2x2 array example step-by-step, watching how results change. This helps understand how to get totals by rows, columns, or entire array.