Tuesday, June 2, 2020

Numpy math functions

Numpy Math Functions

We can perform various mathematical operations upon Numpy ndarrays. There are many functions to carry out trigonometric and arithmetic mathematical functions provided by Numpy. There are many functions available for performing operations on complex numbers.

Trigonometric functions


We can perform several basic and standard functions to perform trigonometric operations. Standard trigonometric functions return trigonometric ratios for given angles in radians.

import numpy as np 
angls = np.array([-30,0,30,45,60,90]) 

print('Sine of different angles:\n') 
# Convert to radians by multiplying with pi/180 
print(np.sin(angls*np.pi/180)) 
print('\n')  

print('Cosine values for angles in array:') 
print(np.cos(angls*np.pi/180)) 
print('\n')  

print('Tangent values for given angles:') 
print(np.tan(angls*np.pi/180)) 

#Output
Sine of different angles:

[-0.5         0.          0.5         0.70710678  0.8660254   1.        ]


Cosine values for angles in array:
[8.66025404e-01 1.00000000e+00 8.66025404e-01 7.07106781e-01
 5.00000000e-01 6.12323400e-17]


Tangent values for given angles:
[-5.77350269e-01  0.00000000e+00  5.77350269e-01  1.00000000e+00
  1.73205081e+00  1.63312394e+16]

arcsin, arcos, and arctan functions return the trigonometric inverse of sin, cos, and tan of the given angle. The result of these functions can be verified by numpy.degrees() function by converting radians to degrees.

import numpy as np 
a = np.array([0,30,45,60,90]) 

print('===========sin values======\n') 
sin_val = np.sin(a*np.pi/180) 
print('Sin of angles',sin_val) 
  

print('=====inverse===============n.') 
inv_of_sin = np.arcsin(sin_val) 
print(inv_of_sin) 

print('=======Check result=======\n') 
print(np.degrees(inv_of_sin)) 

#Output
===========sin values======

Sin of angles [0.         0.5        0.70710678 0.8660254  1.        ]
=====inverse===============n.
[0.         0.52359878 0.78539816 1.04719755 1.57079633]
=======Check result=======

[ 0. 30. 45. 60. 90.]
 

Functions to round off values


There are functions like around(), floor() and ceil() to round off the array items. These methods are applicable on numbers only.

The numpy.around() function


This function returns the rounded value to the desired precision. The syntax of this method is,

numpy.around(arr,decimals)
The arr represents input array and the decimals represent the number of decimals to round off. the default value of this parameter is 0. If it is set to 0, the number is rounded to the left of the decimal position.

import numpy as np 
arr = np.array([3.0,37.55, 168, 1.567, 0.132]) 

print('=====Array arr========\n') 
print(arr)

print('=====rounded arrays======\n') 
print(np.around(arr)) 
print(np.around(arr, decimals = 1)) 
print(np.around(arr, decimals = -1))

#Output
=====Array arr========

[3.000e+00 3.755e+01 1.680e+02 1.567e+00 1.320e-01]
=====rounded arrays======

[  3.  38. 168.   2.   0.]
[3.00e+00 3.76e+01 1.68e+02 1.60e+00 1.00e-01]
[  0.  40. 170.   0.   0.]

The numpy.round_() function

The numpy.round_() function can be used to round an array to the given number of decimals. The syntax of this method is,

numpy.round_(arr, decimals = 0, out = None) 
For example,

import numpy as np

arr = [.50, 1.49, 1.5, 5.5, 6.5, 10.1, 0.558, -6.4,-0.36]
print ("Original array : \n", arr)

round_off_values = np.round_(arr)
print ("\nOutput values : \n", round_off_values)


round_off_values = np.round_(arr, decimals = 1)
print ("\nOutput values : \n", round_off_values)

#Ouput
Original array : 
 [0.5, 1.49, 1.5, 5.5, 6.5, 10.1, 0.558, -6.4, -0.36]

Output values : 
 [ 0.  1.  2.  6.  6. 10.  1. -6. -0.]

Output values : 
 [ 0.5  1.5  1.5  5.5  6.5 10.1  0.6 -6.4 -0.4]



The numpy.floor() function

This method can be used to get the floor value of array items which is the largest integer not greater than the input value.

For example,

import numpy as np

nums=np.array([12.368, 17.8, 2000, 0.368, -12.32, 752.9234])

print(np.floor(nums))

#Output
[  12.   17. 2000.    0.  -13.  752.]

The numpy.ceil() function


This method can be applied to get the ceiling value of the array items which is the smallest integer value greater than the array element.

For example,

import numpy as np

nums=np.array([12.368, 17.8, 2000, 0.368, -12.32, 752.9234])
print("Ceil values of array\n",np.ceil(nums))

#Output
Ceil values of array
[ 1.30e+01  1.80e+01  2.00e+03  1.00e+00 -1.20e+01  7.53e+02]

Other, important functions are

The numpy.rint() round to nearest integer towards zero.
The numpy.fix() round to nearest integer towards zero. 
The numpy.trunc() returns the truncated value of the input, element-wise

logarithmic and Exponent Functions 


There are various functions provided by the Numpy to calculate logarithmic and exponent values of the array items.

For example,

import numpy as np

arr = [2, 3, 4, 5]
print ("Original array : \n", arr)

exp_values = np.exp(arr)
print ("\nExp values : \n", exp_values)

log_values = np.log(arr) #natural log
print ("\nLog values : \n", log_values)


#Output
Original array : 
 [2, 3, 4, 5]

Exp values : 
 [  7.3890561   20.08553692  54.59815003 148.4131591 ]

Log values : 
 [0.69314718 1.09861229 1.38629436 1.60943791]

Other important functions are,

  • The numpy.expm1() function is used to calculate exp(x) – 1 for all elements in the array.
  • The numpy.exp2() function can be used to calculate 2**p for all p in the input array. 
  • The numpy.log2() returns base-2 logarithm of x. 
  • The numpy.log10()function returns the base 10 logarithms of the input array, for each item.
  • The numpy.log1p()function returns the natural logarithm of one plus the input array, element-wise.
  • The numpy.logaddexp() function returns the logarithm of the sum of the exponentiations of the inputs. 
  • The numpy.logaddexp2()function returns the logarithm of the sum of exponentiations of the inputs in base-2.

Numpy Arithmetic  functions


We can apply several arithmetic functions to all the elements in an array to perform general arithmetic operations such as divide(), multiply(), add(), subtract() etc.

import numpy as np
arr1 = [[8, 3, 6],[2, 1, 5],[3, 12, 10]]
arr2 = [[2, 1, 4],[5, 1, 2],[3, 2, 2]]
print("Array 1: \n", arr1)
print("Array 2: \n", arr2)
print("Div = \n", np.divide(arr1, arr2))
print("Mul = \n", np.divide(arr1, arr2))
print("Mod = \n", np.mod(arr1, arr2))

#Output
Array 1: 
 [[8, 3, 6], [2, 1, 5], [3, 12, 10]]
Array 2: 
 [[2, 1, 4], [5, 1, 2], [3, 2, 2]]
Div = 
 [[4.  3.  1.5]
 [0.4 1.  2.5]
 [1.  6.  5. ]]
Mul = 
 [[4.  3.  1.5]
 [0.4 1.  2.5]
 [1.  6.  5. ]]
Mod = 
 [[0 0 2]
 [2 0 1]
 [0 0 0]]

Few important functions are,

  • The numpy.add() is used to add arguments item-wise.
  • The numpy.positive() is used to calculate numerical positive, item-wise.
  • The numpy.negative()is used to calculate numerical negative, item-wise.
  • The numpy.multiply()is used to multiply arguments item-wise.
  • The numpy.power(), method first array items raised to powers from second array, item-wise.
  • The numpy.subtract() is used to calculate subtract arguments, item-wise.
  • The numpy.true_divide() function returns a true division of the inputs, item-wise.
  • The numpy.floor_divide() function returns the largest integer smaller or equal to the division of the inputs.
  • The numpy.float_power() function first array items raised to powers from second array, item-wise.
  • The numpy.mod() function returns the item-wise remainder of division.
  • The numpy.remainder() function return item-wise remainder of division.
  • The numpy.divmod() function return item-wise quotient and remainder simultaneously.


Complex number functions


Numpy offers many functions to work with complex numbers.

import numpy as np 
 
print("check if Real ? ", np.isreal([10+1j, 0j]), "\n")
print("check if Real ? ", np.isreal([10, 15]), "\n")

#Output
check if Real ?  [False  True] 

check if Real ?  [ True  True] 

For example get the conjugate of imaginary numbers,

import numpy as np
 
complx1 = np.array([10+3j, 8+4j])
conj_complx1 = np.conj(complx1)
print ("conjugated complex numbers\n ", conj_complx1)
 
complx2 =np.array([6-7j, 4-3j])
conj_complx2 = np.conj(complx2)
print ("conjugated complex numbers\n", conj_complx2)

#Output
conjugated complex numbers
  [10.-3.j  8.-4.j]
conjugated complex numbers
 [6.+7.j 4.+3.j]

Other mathematical functions


Some other special mathematical functions are,

  • The numpy.convolve() function returns the discrete, linear convolution of two one-dimensional sequences.
  • The numpy.cbrt() function return the non-negative cube-root of an array, item-wise.
  • The numpy.sqrt() function return the non-negative square-root of an array, item-wise.
  • The numpy.square() function return the item-wise square of the input.
  • The numpy.absolute() function calculate the absolute value item-wise.
  • The numpy.fabs() function compute the absolute values item-wise.
  • The numpy.sign() function returns an item-wise indication of the sign of a number.
  • The numpy.interp() function returns one-dimensional linear interpolation.
  • The numpy.maximum() function returns item-wise maximum of array items.
  • The numpy.minimum() function returns item-wise minimum of array items.
  • The numpy.real_if_close(), If complex input returns a real array if complex parts are close to zero.
  • The numpy.nan_to_num() function replaces NaN with zero and infinity with large finite numbers. 
  • The numpy.heaviside() function computes the Heaviside step function.

For example,

import numpy as np
 
nums = np.array([16, 27, 25, 32, 81])
print('Original array\n')
print(nums)
print ("Square root\n ", np.sqrt(nums))
print ("cube root\n ", np.cbrt(nums))
print ("Abs values\n ", np.fabs(np.sqrt(nums)))
print ("Convolve values\n ", np.convolve(np.sqrt(nums),1))

#Output
Original array

[16 27 25 32 81]
Square root
  [4.         5.19615242 5.         5.65685425 9.        ]
cube root
  [2.5198421  3.         2.92401774 3.1748021  4.32674871]
Abs values
  [4.         5.19615242 5.         5.65685425 9.        ]
Convolve values
  [4.         5.19615242 5.         5.65685425 9.        ]