Python’s NumPy library is fun in that it’s easy to work with multi-dimensional data.  For simplicity, consider a 2D array (aka matrix).

I wrote some code to demonstrate the creation, simple visualization, slicing, and aggregation of data within a matrix, including totals and slice-subtotals.

Source Code:

It is available in Git Hub: NumPy 2D Array Slice Aggregation.

Here we go…

Step-by-Step Demonstration of Python Code and Results

# Create matrix
matrix = np.arange(6,31).reshape(5,5)
matrix

# Results: Note that two level [ [ … ] ] brackets means two dimensional array (metrix)
#array([[ 6, 7, 8, 9, 10],
#            [11, 12, 13, 14, 15],
#            [16, 17, 18, 19, 20],
#            [21, 22, 23, 24, 25],
#            [26, 27, 28, 29, 30]])
#
#
# Find mean (average) value for each column.
matrix.mean(axis=0)

# Results…
#array([16., 17., 18., 19., 20.])
#
#
# Now, mean values per row, and reshape for vertical output
matrix.mean(axis=1).reshape(5,1)

# Results:
#array([[ 8.],
#            [13.],
#            [18.],
#            [23.],
#            [28.]])
#
#
# Slice matrix just to rows 1 – 3, columns 2 & 3
matrix[:3,1:3]

# Results…
#array([[ 7, 8],
#            [12, 13],
#            [17, 18]])
#
#
#Sum the slice on each column
matrix[:3,1:3].sum(axis=0)

# Results…
#array([36, 39])
#
#
#Sum the entire slice
matrix[:3,1:3].sum()

# Results…
#75
#
#
#Sum the slice on each row, and reshape
matrix[:3,1:3].sum(axis=1).reshape(3,1)

# Results…
#array([[15],
#             [25],
#             [35]])
#
#
#As above, sum the whole slice as visual check
# that row and column aggregates add up.
matrix[:3,1:3].sum()

# Results: Same as above
#75
#
#

# # # All done! # # #

As we’ve seen here, a grasp of NumPy fundamentals makes creating, slicing, and aggregating multi-dimensional data straightforward.

With that, my journey continues into Python Pandas.