In this post, we will learn about different data types in numpy and also we will see how we can convert data types of numpy array with detailed explanations and examples.

You can also check out our previous article πŸ‘‰ Create a NumPy Array with Random Values

So, let’s start learning. First, we will explore all the data types of NumPy, and after that, we will try them with examples.

Data Types in NumPy

Data TypesDescription
bool_Boolean (True or False) stored as a byte.
int_Default integer type normally either int64 or int32
intcIdentical to C int (normally either int32 or int64)
intpInteger used for indexing (normally either int32 or int64)
int88-bit integer (-128 to 127)
int1616-bit integer (-32,768 to 32,767)
int3232-bit integer (-2,147,483,648 to 2,147,483,647)
int6464-bit integer (-9,223,372,036,854,775,808 to 9,223,372,036,854,775,807)
uint88-bit unsigned integer (0 to 255)
uint1616-bit unsigned integer (0 to 65,535)
uint3232-bit unsigned integer (0 to 4,294,967,295)
uint6464-bit unsigned integer (0 to 18,446,744,073,709,551,615)
float_Short-hand for float64
float1616-bit floating-point number; Half precision floating-point; Sign (1 bit), Exponent (5 bits), Mantissa (10 bits)
float3232-bit floating-point number; Single precision floating-point; Sign (1 bit), Exponent (8 bits), Mantissa (23 bits)
float6464-bit floating-point number; Double precision floating-point; Sign (1 bit), Exponent (11 bits), Mantissa (52 bits)
complex_Short-hand for complex128
complex64Complex number represented by two 32-bit floats (Real and imaginary component)
complex128Complex number represented by two 64-bit floats (Real and imaginary component)
string_Fixed-size ASCII strings
timedelta64timedelta type for representing time intervals
datetime64Date and time with nanosecond precision.
voidVoid type used for structured arrays.

Numpy also provides a feature to represent all the data types using their special characters. So let’s explore all the special characters used to represent data types in NumPy.

List of Characters that are Used to Represent Data Types in NumPy

  • i – integer
  • b – boolean
  • u – unsigned integer
  • f – float
  • c – complex float
  • m – timedelta
  • M – datatime
  • O – object
  • S – String
  • U – Unicode String
  • V – void (fixed chunk of memory for other types)

Check data types of NumPy Array

To check the data type of a numpy array, you can use the dtype attribute.

import numpy as np

arr = np.array([1,2,3,4,5])



Create a NumPy Array with Specified Data Types

To create a NumPy array with a specified data type, you need to pass the dtype attribute as an argument inside the array() function.

import numpy as np

arr2 = np.array([1,2,3,4,5], dtype="int8")


[1 2 3 4]

In NumPy, each data type has a corresponding built-in function, such as numpy.int8() for int8. You can also use these built-in functions in the dtype attribute to create arrays with specified data types.

import numpy as np

arr3 = np.array([2,4,6,8,10], dtype=np.int8)


[ 2  4  6  8 10]

There is one more way to specify the dtype, which is to use special characters for data types.

import numpy as np

arr4 = np.array([1,2,3,4,5,6,7], dtype="f")


[1. 2. 3. 4. 5. 6. 7.]

Convert Data Types of a Numpy Array

There is two ways to convert data types of an array:

  1. Using built-in functions &
  2. Using astype() method

Using Built-in Function

As we have already discussed, each data type in NumPy has a corresponding function. We can use these functions to convert one data type into another.

Int β†’ Float

import numpy as np

a1_int = np.array([1,2,3,5,7,11,13,17])
print(a1_int.dtype) # Output: int32


# Convert int --> float

a1_float = np.float16(a1_int)
print(a1_float.dtype) # Output: float16


[ 1  2  3  5  7 11 13 17]

[ 1.  2.  3.  5.  7. 11. 13. 17.]

Float β†’ Int

import numpy as np

a2_float = np.array([1,2,3,5,7,11,13,17], dtype="float32")
print(a2_float.dtype) # Output: float64


# Convert float --> int
a2_int = np.int16(a2_float)


[ 1.  2.  3.  5.  7. 11. 13. 17.]

[ 1  2  3  5  7 11 13 17]

Using astype() method

The astype() function used to convert data types of an array. It takes the target data type as an argument, and it returns a new array with the same data but in the specified data type.

arr = np.array([1,2,3,4,5])
new_arr = arr.astype("float32")
print(new_arr.dtype) # Output: float32


[1. 2. 3. 4. 5.]

Hi, I'm Yagyavendra Tiwari, a computer engineer with a strong passion for programming. I'm excited to share my programming knowledge with everyone here and help educate others in this field.

Write A Comment