Determine common type following standard coercion rules in Python

In NumPy, find_common_type() determines the common data type following standard coercion rules. This function helps when working with mixed data types in arrays and scalars, returning the most appropriate common type.

Syntax

numpy.find_common_type(array_types, scalar_types)

Parameters

The function takes two parameters:

• array_types − A list of dtypes or dtype convertible objects representing arrays
• scalar_types − A list of dtypes or dtype convertible objects representing scalars

How It Works

The method returns the common data type, which is the maximum of array_types ignoring scalar_types, unless the maximum of scalar_types is of a different kind (dtype.kind). If the kind is not understood, then None is returned.

Example

Let's explore various combinations of array and scalar types ?

import numpy as np

print("Using the find_common_type() method in NumPy\n")

# Float32 array with int64 and float64 scalars
result1 = np.find_common_type([np.float32], [np.int64, np.float64])
print("Float32 array with int64/float64 scalars:", result1)

# Empty array types with mixed scalars
result2 = np.find_common_type([], [np.int64, np.float32, complex])
print("Empty array with mixed scalars:", result2)

# Float32 array with complex scalar
result3 = np.find_common_type([np.float32], [complex])
print("Float32 array with complex scalar:", result3)

# Float64 array with complex scalar
result4 = np.find_common_type([np.float64], [complex])
print("Float64 array with complex scalar:", result4)

# String dtype representations
result5 = np.find_common_type(['f4', 'i4'], ['c8'])
print("String dtypes f4/i4 arrays with c8 scalar:", result5)

# Integer array with float scalar
result6 = np.find_common_type([np.int64], [np.float64])
print("Int64 array with float64 scalar:", result6)

Using the find_common_type() method in NumPy

Float32 array with int64/float64 scalars: float32
Empty array with mixed scalars: complex128
Float32 array with complex scalar: complex128
Float64 array with complex scalar: complex128
String dtypes f4/i4 arrays with c8 scalar: complex128
Int64 array with float64 scalar: float64

Key Points

• Array types take precedence over scalar types in determining the common type
• When array types are empty, the function considers only scalar types
• Complex types generally dominate other numeric types
• The function follows NumPy's standard type promotion rules

Conclusion

The find_common_type() function is essential for determining compatible data types when working with mixed NumPy arrays and scalars. It follows standard coercion rules to ensure type safety in numerical computations.