HumachLearnLearn · explore · apply
Start learning
NumPy for Numerical Computing · Lesson 107

Vectorised Arithmetic

Vectorised Arithmetic is part of NumPy's array model.

ConceptWorked examplePracticeKnowledge check
Textbook walkthrough

What Vectorised Arithmetic actually means

Vectorised Arithmetic is part of NumPy's array model. An ndarray stores homogeneous values in an n-dimensional shape and enables vectorised operations that act over many elements without writing an explicit Python loop for each value.

Vectorised Arithmetic matters because NumPy represents numerical data as shaped arrays and performs operations across whole axes efficiently. Shape, dtype, broadcasting and reduction rules directly determine whether a numerical result means what you think it means.

Deeper walkthrough

Read Vectorised Arithmetic as a mechanism, not a recipe

Treat this as a sequence of observable decisions rather than one opaque command. Stage 1: Create an ndarray and inspect shape, ndim and dtype. Stage 2: Select values with indexing, slicing or boolean masks. Stage 3: Apply vectorised arithmetic; broadcasting aligns compatible dimensions. Final checkpoint: Treat NaN deliberately because many aggregations either propagate it or require nan-aware functions.

Mechanism

Follow the transformation

Create an ndarray and inspect shape, ndim and dtype.

Select values with indexing, slicing or boolean masks.

Apply vectorised arithmetic; broadcasting aligns compatible dimensions.

Evidence

Know what would convince you

  • Inspect shape, ndim and dtype before and after the operation.
  • Hand-check one slice/row/column or tiny numeric case against the vectorised result.
Useful distinctionPython list: General-purpose heterogeneous container.
Click a stage to inspect what happens, what changes, and what should be checked before moving on.
Stage 1

Create an ndarray and inspect shape

Create an ndarray and inspect shape, ndim and dtype. For Vectorised Arithmetic, make this checkpoint explicit by recording the evidence inspected, the expected result, and the condition that would make you reject the current result.

Verification focus: record the evidence you inspected and the condition that would make this stage fail.
How it works

Trace the mechanism step by step

  1. Create an ndarray and inspect shape, ndim and dtype.
  2. Select values with indexing, slicing or boolean masks.
  3. Apply vectorised arithmetic; broadcasting aligns compatible dimensions.
  4. Reshape only when the number of elements remains consistent.
  5. Treat NaN deliberately because many aggregations either propagate it or require nan-aware functions.
Worked demonstration

Make the concept concrete

Demonstration

Python / NumPy example

# Step 1 — Import the module so its functions/classes are available to the rest of this example.
import numpy as np
# Step 2 — Construct `X` as an array so vectorised numerical operations can be applied consistently.
X = np.array([[1., 2., 3.], [4., 5., 6.]])
# Step 3 — Compute the right-hand expression and store its result in `centre` for the next step.
centre = X.mean(axis=0)
# Step 4 — Compute the right-hand expression and store its result in `Xc` for the next step.
Xc = X - centre          # broadcasting over rows
# Step 5 — Display the current value explicitly so the result/state can be inspected during execution.
print("shape:", X.shape)
# Step 6 — Display the current value explicitly so the result/state can be inspected during execution.
print("column means:", centre)
# Step 7 — Display the current value explicitly so the result/state can be inspected during execution.
print("centred:\n", Xc)
Expected / illustrative result
shape: (2, 3)
column means: [2.5 3.5 4.5]
centred:
 [[-1.5 -1.5 -1.5]
 [ 1.5  1.5  1.5]]
Interpret the result.

For Vectorised Arithmetic, connect the displayed result to the specific input and mechanism above; independently verify one value/state change rather than treating successful execution as proof.

Distinctions & related ideas

Know what this is — and what it is not

Python listGeneral-purpose heterogeneous container.
NumPy ndarrayHomogeneous n-D numerical array with vectorised operations.
BroadcastingImplicitly expands size-1/missing dimensions when shapes are compatible.
ReshapeChanges the array view/shape without changing element count.
Use deliberately

When it is appropriate

Use Vectorised Arithmetic when the task benefits from homogeneous n-dimensional arrays, vectorised numerical operations or explicit shape/dtype semantics.

Boundary conditions

When to stop or reconsider

Pause when shapes are unclear, dtypes would lose information, broadcasting is accidental, or labelled/tabular semantics would be safer in pandas.

Common mistakes

Failure modes to recognise

  • Ignoring shape or axis and obtaining a numerically valid result along the wrong dimension.
  • Allowing dtype conversion or integer/float behaviour to silently change precision or missing-value handling.
  • Assuming broadcasting matches intent without writing down the aligned dimensions.
Verification

How to check the result

  • Inspect shape, ndim and dtype before and after the operation.
  • Hand-check one slice/row/column or tiny numeric case against the vectorised result.
  • For broadcasting or reshaping, write the expected resulting shape before execution and confirm it explicitly.
Hands-on practice

Demonstrate understanding

Try this:

Build a tiny, inspectable example of Vectorised Arithmetic. First create an ndarray and inspect shape, ndim and dtype. Then select values with indexing, slicing or boolean masks. Write the expected result before running it, and explain one condition that would make the result misleading or invalid.

Use a 2×3 or similarly tiny array. Keep shape and dtype visible and verify one element/axis calculation by hand.
Knowledge check

Check reasoning, not memorisation

Before trusting a result from Vectorised Arithmetic, which check provides the strongest evidence that you understand and applied it correctly?

Quick reference

Keep the important distinctions visible

Step 1Create an ndarray and inspect shape, ndim and dtype.
Step 2Select values with indexing, slicing or boolean masks.
Step 3Apply vectorised arithmetic; broadcasting aligns compatible dimensions.
Step 4Reshape only when the number of elements remains consistent.
Lesson summary

What to remember

  • Vectorised Arithmetic is part of NumPy's array model. An ndarray stores homogeneous values in an n-dimensional shape and enables vectorised operations that act over many elements without writing an explicit Python loop for each value.
  • Create an ndarray and inspect shape, ndim and dtype.
  • Ignoring shape or axis and obtaining a numerically valid result along the wrong dimension.
  • Inspect shape, ndim and dtype before and after the operation.