Data type Object (dtype) in NumPy Python

Every ndarray has an associated data type (dtype) object. This data type object (dtype) informs us about the layout of the array. This means it gives us information about :

• Type of the data (integer, float, Python object etc.)
• Size of the data (number of bytes)
• Byte order of the data (little-endian or big-endian)
• If the data type is a sub-array, what is its shape and data type.

The values of an ndarray are stored in a buffer which can be thought of as a contiguous block of memory bytes. So how these bytes will be interpreted is given by the dtype object.  

1. Constructing a data type (dtype) object : Data type object is an instance of numpy.dtype class and it can be created using numpy.dtype.

   Parameters:

   • obj: Object to be converted to a data type object.
   • align : bool, optional
     Add padding to the fields to match what a C compiler would output for a similar C-struct.
   • copy : bool, optional
     Make a new copy of the data-type object. If False, the result may just be a reference to a built-in data-type object.

   # Python Program to create a data type object import numpy as np    # np.int16 is converted into a data type object. print(np.dtype(np.int16))

   Output:

   
   
   

   int16

   # Python Program to create a data type object  # containing a 32 bit big-endian integer import numpy as np    # i4 represents integer of size 4 byte # > represents big-endian byte ordering and < represents little-endian encoding. # dt is a dtype object dt = np.dtype('>i4')    print("Byte order is:",dt.byteorder)    print("Size is:",dt.itemsize)    print("Data type is:",dt.name)

   Output:

   Byte order is: >
   Size is: 4
   Name of data type is: int32
   

   The type specifier (i4 in above case) can take different forms:

   1. b1, i1, i2, i4, i8, u1, u2, u4, u8, f2, f4, f8, c8, c16, a
         (representing bytes, ints, unsigned ints, floats, complex and
          fixed length strings of specified byte lengths)
   2. int8,…,uint8,…,float16, float32, float64, complex64, complex128
         (this time with bit sizes)

   Note :

   dtype is different from type. 
   

   # Python program to differentiate # between type and dtype. import numpy as np    a = np.array([1])    print("type is: ",type(a)) print("dtype is: ",a.dtype)

   Output:

   type is:    
   dtype is:  int32
   

2. Data type Objects with Structured Arrays : Data type objects are useful for creating structured arrays.  A structured array is the one which contains different types of data. Structured arrays can be accessed with the help of fields.
   A field is like specifying a name to the object. In case of structured arrays the dtype object will also be structured. 

   # Python program for demonstrating  # the use of fields import numpy as np    # A structured data type containing a 16-character string (in field ‘name’)  # and a sub-array of two 64-bit floating-point number (in field ‘grades’):    dt = np.dtype([('name', np.unicode_, 16), ('grades', np.float64, (2,))])    # Data type of object with field grades print(dt['grades'])    # Data type of object with field name  print(dt['name'])

   Output:

   ('<f8', (2,))
   
   

   # Python program to demonstrate  # the use of data type object with structured array. import numpy as np    dt = np.dtype([('name', np.unicode_, 16), ('grades', np.float64, (2,))])    # x is a structured array with names and marks of students. # Data type of name of the student is np.unicode_ and  # data type of marks is np.float(64) x = np.array([('Sarah', (8.0, 7.0)), ('John', (6.0, 7.0))], dtype=dt)    print(x[1]) print("Grades of John are: ",x[1]['grades']) print("Names are: ",x['name'])

   
   Output:

   ('John', [ 6.,  7.])
   Grades of John are:  [ 6.  7.]
   Names are:  ['Sarah' 'John']
   

References :

• docs.scipy.org
• Structured Arrays

Please write comments if you find anything incorrect, or you want to share more information about the topic discussed above.