NumPy ndarray & Array Creation Methods

Python tutorial · PySpark.in

What is ndarray?

An ndarray (N-dimensional array) is a fundamental data structure in NumPy that stores elements of the same data type in a multi-dimensional, fixed-size array, enabling fast and memory-efficient numerical computations.

Creating Arrays using array()

Syntax

```

Hide

np.array(object, dtype=None)

```

Example

```

import numpy as np

arr = np.array([1, 2, 3, 4])

print(arr)

```

1D, 2D, and 3D Arrays

1D Array

```

a = np.array([10, 20, 30])

print(a)

```

Used for: lists, vectors, simple data

2D Array

```

b = np.array([[1, 2], [3, 4]])

print(b)

```

Used for: matrices, tables, datasets

3D Array

```

c = np.array([

[[1, 2], [3, 4]],

[[5, 6], [7, 8]]

])

print(c)

```

Used for: images, videos, ML tensors

Array Attributes

ndim – Number of Dimensions

```

A=a.ndim

print(A)

```

Tells whether the array is 1D, 2D, or 3D.

shape – Structure of Array

```

B=b.shape

print(B)

```

Represents rows × columns.

size – Total Elements

```

B=b.size

print(B)

```

Total number of elements in the array.

Data Type of Array

dtype

```

X=arr.dtype

print(X)

```

Shows type of elements stored.

astype() – Type Conversion

```

arr_float = arr.astype(float)

print(arr_float)

```

Creates a new array with changed data type.

Memory Size of Array

itemsize

```

X=arr.itemsize

print(X)

```

Shows memory (in bytes) used by one element.

Array Creation Methods

NumPy provides built-in functions to create arrays efficiently without manual data entry.

1.zeros()

Creates an array filled with zeros.

```

P=np.zeros((3, 3))

print(P)

```

Used for initialization in ML & matrices.

2.ones()

Creates an array filled with ones.

```

K=np.ones((2, 2))

print(K)

```

Useful in normalization and bias initialization.

3.empty()

Creates an array without initializing values.

```

S=np.empty((2, 2))

print(S)

```

Faster but contains garbage values.

4.full()

Creates an array filled with a specific value.

```

Y=np.full((2, 3), 7)

print(Y)

```

Used when constant values are needed.

5.eye(){identity matrix}:Used in linear algebra & matrix operations.

```

X=np.eye(3)

print(X)

```

6.arange()

Creates array with a range of values.

```

X=np.arange(0, 10, 2)

print(X)

```

Similar to range() but returns NumPy array.

7.linspace()

Creates array with evenly spaced values.

8.np.linspace(0, 1, 5)

Used in graphs, ML scaling, statistics.

Practical Example (Combined)

```

import numpy as np

a = np.zeros((2,2))

b = np.ones((2,2))

c = np.eye(2)

d = np.arange(1, 10)

e = np.linspace(1, 5, 5)

print(a, b, c, d, e)

```

More Python tutorials

All tutorials · Try the free PySpark compiler · Practice challenges