Data Analysis - Day8 - Numpy Review1

The difference between a Python list and a Numpy array

1. Numpy gives you an enormous range of fast and efficient ways of creating arrays and manipulating numerical data inside them.
2. Python list can contain different data types within a single list
3. Numpy array of all the elements should be homogeneous.

Creation

Python list creation

Python list

list01 = [1.4,2,'3',[3,5,6],{2,3,4}]
print(list01)
print(type(list01))
name = 'HHVic'
print(list(name))

[1.4, 2, '3', [3, 5, 6], {2, 3, 4}]
<class 'list'>
['H', 'H', 'V', 'i', 'c']

Numpy array creation

Numpy
numpy.array(object, dtype = None, copy = True, order = None, subok = False, ndmin = 0)

名称	描述
object	Array or Nested array
dtype	The data type of the array element, optional
copy	The object needs to be copied, or not. optional
order	Create array style, C for row, F for column, A for any direction(default)
subok	Return an array consistent with the base class type by default
ndmin	Specifies the minimum dimension of the array

array01 = np.arange(1,10,2)   #(start, stop, step, dtype)
print(array01)
print(type(array01))
name1 = 'HHVic'
print(np.array(name1))

[1 3 5 7 9]
<class 'numpy.ndarray'>
HHVic

Create 1 dimension and 2 dimension array

1. Assignment directly

array1d = np.array([1,2,3,4])
print(array1d)
array2d= np.array([[1,2],[3,4],[5,6]])
print(array2d)
print(type(array2d))

[1 2 3 4]
[[1 2]
 [3 4]
 [5 6]]
<class 'numpy.ndarray'>

2. USE reshape()

arrayreshape = np.arange(10).reshape(2,5,order ='C')
print(arrayreshape)
arrayreshape = np.arange(10).reshape(2,5,order ='F')
print(arrayreshape)

[[0 1 2 3 4]
 [5 6 7 8 9]]
[[0 2 4 6 8]
 [1 3 5 7 9]]

3. Use randint()

arrayrandint = np.random.randint(0,10,(4,4))
print(arrayrandint)

[[3 5 6 1]
 [8 7 7 0]
 [6 8 3 9]
 [0 2 3 7]]

More Numpy array functions

1. Zeros

zeros= np.zeros(11, dtype = float, order ='C')
print(zeros)

[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]

2. Ones

ones = np.ones(10, dtype = float, order ='C')
print(ones)

[1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]

3. linspace()

linspaces = np.linspace(0,5,5)
print(linspaces)

[0.   1.25 2.5  3.75 5.  ]

4. random()

randoms = np.random.rand(2,3)
print(randoms)
randoms = np.random.randint(2,10,size=10) # produce 10 random number form 2 to 9
print(randoms)
randoms = np.random.randn(5)
print(randoms)

[[0.48395755 0.60356083 0.82227284]
 [0.39616543 0.11712723 0.23434582]]
[5 7 4 9 3 8 6 4 3 7]
[ 0.74630555  0.8983705   0.96306982  1.63266755 -1.72612846]

5. empty()

emptys = np.empty((2,5),dtype=int)
print(emptys)

[[ 727395495 1072161212 2064842652 1072480115 -857168134]
 [1072615799   28942093 1073356648 -527735117 1073454648]]

Adding, Indexing , Assigning , Slicing, Removing

Python list

Python list

#Adding
list01 =[1.4, 2, 'SSS', '3', [3, 5, 6], {2, 3, 4}]
list01.append('abcd')
print(list01)
list01.insert(0,'ABCD')
print(list01)
list01.extend('extend1')   #(sequence)
print(list01)

list01 =[1.4, 2, 'SSS', '3', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'H', 'V', 'i', 'c']

#Indexing and Slicing
print(list01[5])
print(list01[-1]) # The last element
print(list01[:-3])


#Assigning
list01[3]='Dream'  #list[index] = elements
print(list01)

list01[1:3] = ('My','New')   #list[slicing] = container
print(list01)

list02 = list01[3:9]
print(list02)

# Removing
del list01[8]  #Remove the 8th index == 'H'
print(list01)
list01.remove('V')  #(element)Remove the 1st element after search from first to end
print(list01)
list01.pop(9)  # (index) Remove the value of the index, if no index, remove the last element
print(list01)

#list01.append('abcd')
[1.4, 2, 'SSS', '3', [3, 5, 6], {2, 3, 4}, 'abcd']
#list01.insert(0,'ABCD')
['ABCD', 1.4, 2, 'SSS', '3', [3, 5, 6], {2, 3, 4}, 'abcd']
#list01.extend('extend1')  
['ABCD', 1.4, 2, 'SSS', '3', [3, 5, 6], {2, 3, 4}, 'abcd', 'e', 'x', 't', 'e', 'n', 'd', '1']

#print(list01[5])
{2, 3, 4}
#print(list01[-1]) 
c
#print(list01[:-3])
[1.4, 2, 'SSS', '3', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'H']
#list01[3]='Dream' 
[1.4, 2, 'SSS', 'Dream', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'H', 'V', 'i', 'c']
#list01[1:3] = ('My','New') 
[1.4, 'My', 'New', 'Dream', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'H', 'V', 'i', 'c']
#list02 = list01[3:9]
['Dream', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'H']
#del list01[8] 
[1.4, 'My', 'New', 'Dream', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'V', 'i', 'c']
#list01.remove('V') 
[1.4, 'My', 'New', 'Dream', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'i', 'c']
#list01.pop(9) 
[1.4, 'My', 'New', 'Dream', [3, 5, 6], {2, 3, 4}, 'HHVic', 'H', 'i']

Numpy array

Numpy array

Adding

#Adding
#numpy.append(arr, values, axis=None)

array01 = np.array([[1,2,3],[4,5,6]])
print(array01)
array02 = np.append(array01,[[7,8,9]]) # array will be replaced to 1d array
print(array02)
print(type(array02))
array03 = np.append(array01,[[7,8,9]],axis=0) # axis =0 add to row
print(array03)
array04 = np.append(array01,[[5,5,5],[7,8,9]],axis=1) # axis =1 add to column
print(array04)

[[1 2 3]
 [4 5 6]]
[1 2 3 4 5 6 7 8 9]
<class 'numpy.ndarray'>
[[1 2 3]
 [4 5 6]
 [7 8 9]]
[[1 2 3 5 5 5]
 [4 5 6 7 8 9]]

#numpy.insert(arr, obj, values, axis)
array01 = np.array([[1,2,3],[4,5,6]])
print(array01)
array02 = np.insert(array01,2,[7,8,9])  #insert to the location of index 2
print(array02)
array03 = np.insert(array01,2,[7,8,9],axis=0)
print(array03)
array04 = np.insert(array01,2,[7,8],axis=1)
print(array04)

[[1 2 3]
 [4 5 6]]
[1 2 7 8 9 3 4 5 6]
[[1 2 3]
 [4 5 6]
 [7 8 9]]
[[1 2 7 3]
 [4 5 8 6]]

Indexing and slicing (1 dimension)

# 1 dimension indexing and slicing
array1d = np.arange(1,8)
print(array1d)   # [1 2 3 4 5 6 7]
print(array1d[5:])   #6 7
print(array1d[:5])   #1 2 3 4 5
print(array1d[:-5:-1]) #7 6 5 4
print(array1d[-5::-1]) #3 2 1
print(array1d[::]) #1 2 3 4 5 6 7
print(array1d[:])  #1 2 3 4 5 6 7

Indexing and slicing (multiple dimensions)

#array[page,row_index,column_index]
array3d = np.arange(24).reshape(2,3,4)
print(array3d)

print('[1,:,:]:\n', array3d[1,:,:])  # all elements in page 1

print('[:,1,:]:\n', array3d[:,1,:])  # all 1st rows in all pages

print('[1,:,1]:\n', array3d[1,:,1])  # the 1st columns in page 1

print('*'*80)

[[[ 0  1  2  3]
  [ 4  5  6  7]
  [ 8  9 10 11]]
 [[12 13 14 15]
  [16 17 18 19]
  [20 21 22 23]]]
[1,:,:]:
 [[12 13 14 15]
 [16 17 18 19]
 [20 21 22 23]]
[:,1,:]:
 [[ 4  5  6  7]
 [16 17 18 19]]
[1,:,1]:
 [13 17 21]
 ********************************************************************************

Assigning

array3d = np.arange(1,9).reshape(2,2,2)
print(array3d)
array3d[1,:,:] = 0
print(array3d)

[[[1 2]
  [3 4]]

 [[5 6]
  [7 8]]]
[[[1 2]
  [3 4]]

 [[0 0]
  [0 0]]]

Use condition

array3d = np.arange(1,9).reshape(2,2,2)
print(array3d)
array3d[array3d < 5] = 0
print(array3d)

[[[0 0]
  [0 0]]

 [[5 6]
  [7 8]]]

Removing and deleting duplicate elements

numpy.delete(arr, obj, axis)

array3d = np.arange(1,9).reshape(2,2,2)
print(array3d)
array3d_del= np.delete(array3d,1,axis=0)
print(array3d_del)

[[[1 2]
  [3 4]]

 [[5 6]
  [7 8]]]
[[[1 2]
  [3 4]]]

numpy.unique(arr, return_index, return_inverse, return_counts)

1-Dimension

array1d = np.arange(10)
array1d = np.append(array1d,[5,6,7,8,9])
print(array1d)
array1d = np.unique(array1d)
print(array1d)

[0 1 2 3 4 5 6 7 8 9 5 6 7 8 9]
[0 1 2 3 4 5 6 7 8 9]

Multiple Dimension

array1d = np.array([3,4,7,3,8,2,6,3,4,7,9,1,5,8,2])
print('array1d:',array1d)
array1dnew = np.unique(array1d)
print('array1d_unique:',array1dnew)
a1, indices_a1 = np.unique(array1d, return_index=True)
print(a1) # print the unique array, and sorted from small to big
print(indices_a1) # put the old unique elements' index into a array.
a2, indices_a2 = np.unique(array1d, return_inverse=True)
print(a2) # same as a1
print(indices_a2)  # get the index of all old elements and sorted one bye one
a3, indices_a3 = np.unique(array1d, return_counts=True)
print(a3) # same as a1
print(indices_a3)

array1d: [3 4 7 3 8 2 6 3 4 7 9 1 5 8 2]
array1d_unique: [1 2 3 4 5 6 7 8 9]
[1 2 3 4 5 6 7 8 9]
[11  5  0  1 12  6  2  4 10]
[1 2 3 4 5 6 7 8 9]
[2 3 6 2 7 1 5 2 3 6 8 0 4 7 1]
[1 2 3 4 5 6 7 8 9]
[1 2 3 2 1 1 2 2 1]