Machine Learning Summary - NumPy basics

model.py

…

Expected output

Lesson notes

Machine Learning

Summary - NumPy basics

In this group you learned to create NumPy arrays, use their properties, and index them.

Summary: NumPy basics

What you learned in this group:

Creating arrays with np.array, arange, linspace, zeros.
Inspecting arrays with shape, ndim, size, dtype.
Selecting with a[row, column], slices and boolean masks.
Changing shape with reshape, flatten, .T.

X = np.arange(12).reshape(4, 3)
print(X[:, 0])          # first feature
print(X[X[:, 0] > 3])   # rows whose first feature is above 3

Element-wise operations and vectorisation are the key to fast, short ML code.

Exercise All together

Turn the numbers 0-11 into a 4×3 matrix. Select the last column, filter the values above 5 with a boolean mask and print them.

Show example solution

import numpy as np

X = np.arange(12).reshape(4, 3)
son = X[:, -1]
print(son[son > 5])

Test yourself

What does <code>np.arange(12).reshape(4, 3)[:, 0]</code> give?
- 0
- [0 3 6 9]
- [0 1 2]
- [0 4 8]
Show answer

[0 3 6 9] — The rows are [0 1 2], [3 4 5], [6 7 8], [9 10 11]; the first element of each is taken.
Compute the sum of all elements.
```
toplam = X.____()
```
Show answer
```
toplam = X.sum()
```
Match the tools with their roles.

linspace

Evenly spaced values
astype

Type conversion
reshape

Changing shape
x[x > 0]

Conditional selection

Next lesson: What is Pandas?