Day 3. Numpy01

#!/usr/bin/env python
# coding: utf-8
 
import numpy as np
 
np_array = np.array([3, 33, 333]) #create rank1 array
 
print(type(np_array))
print(np_array)
print(np_array.shape)
print(np_array[0], np_array[1], np_array[2])
print(np_array)
another = np.array([[11,12,13], 
                    [21,22,23]])
 
print(another)
print("shape of another {}".format(another.shape))
 
another[0][2]
np.zeros((3,3))
 
np.full((3,3), 9.0)
 
np.eye(3,3)
 
np.ones((1,2))
 
np.ones((1,2)).shape  #rank 2 matrix
 
np.random.random((2,2))
 
# Rank 2 array of shape (3, 4)
an_array = np.array([[11,12,13,14], [21,22,23,24], [31,32,33,34]])
print(an_array)
 
a_slice = an_array[:2, 1:3]  # copy by referrence
a_slice[0,0] = 123
print(a_slice)
 
a_slice = np.array(an_array[:2, 1:3]) #call by value
a_slice[0,0] = 321
print(a_slice)
 
row_rank1 = an_array[1, :]
print(row_rank1, row_rank1.shape)
 
row_rank2 = an_array[1:2, :]    #this gives 2d array
print(row_rank2, row_rank2.shape)
 
an_array = np.array([[11,12,13], [21,22,23], [31,32,33],[41,42,43] ])
 
an_array
 
col_indices = np.array([0, 1, 2, 0])
print(col_indices)
    
row_indices = np.arange(4)
print(row_indices)
 
#Examine the pairings of row_indices and col_indices.
for row, col in zip( row_indices, col_indices):
    print(row, col)
 
# Select one element from each row
an_array[row_indices, col_indices]
 
an_array[row_indices, col_indices] += 100000
an_array
 
an_array = np.array([[11,12], [21,22], [31, 32]])
 
an_array
 
# boolean values will be returned as a list
filter = (an_array > 15)
filter
 
an_array[filter]
 
#better way
an_array[an_array > 15]
 
an_array[an_array > 25] = 25 
an_array
 
## Data type of an element in array is matter!!
ex1 = np.array([11, 12])
ex1.dtype
 
ex2 = np.array([11.0, 12.0])
ex2.dtype
 
ex3 = np.array([11,12], dtype=np.int64)
ex3.dtype
 
ex4 = np.array([11.1, 12.7], dtype = np.int64)
ex4.dtype
 
ex5 = np.array([ 11, 12 ], dtype = np.int64)
ex5.dtype
ex5
 
## ndarray operation
 
x = np.array([[ 111,112     ],     [121,122 ]], dtype = np.int)
y = np.array([[ 211.1,212.1 ], [221.1,222.1 ]], dtype = np.int)
print(x)
print()
print(y)
 
x + y
 
np.add(x,y)
 
np.sqrt(x)
 
np.exp(x)
 

import numpy as np

np_array = np.array([3, 33, 333]) #create rank1 array

print(type(np_array))
print(np_array)

<class 'numpy.ndarray'>
[  3  33 333]

In [18]:

print(np_array.shape)

(3,)

In [19]:

print(np_array[0], np_array[1], np_array[2])

3 33 333

In [20]:

print(np_array)

[  3  33 333]

In [24]:

another = np.array([[11,12,13], 
                    [21,22,23]])
print(another)
print("shape of another {}".format(another.shape))

[[11 12 13]
 [21 22 23]]
shape of another (2, 3)

In [26]:

another[0][2]

Out[26]:

13

In [31]:

np.zeros((3,3))

Out[31]:

array([[0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.]])

In [32]:

np.full((3,3), 9.0)

Out[32]:

array([[9., 9., 9.],
       [9., 9., 9.],
       [9., 9., 9.]])

In [33]:

np.eye(3,3)

Out[33]:

array([[1., 0., 0.],
       [0., 1., 0.],
       [0., 0., 1.]])

In [35]:

np.ones((1,2))

Out[35]:

array([[1., 1.]])

In [36]:

np.ones((1,2)).shape  #rank 2 matrix

Out[36]:

(1, 2)

In [37]:

np.random.random((2,2))

Out[37]:

array([[0.1570544 , 0.06688494],
       [0.25285217, 0.54726879]])

In [3]:

# Rank 2 array of shape (3, 4)
an_array = np.array([[11,12,13,14], [21,22,23,24], [31,32,33,34]])
print(an_array)

[[11 12 13 14]
 [21 22 23 24]
 [31 32 33 34]]

In [44]:

a_slice = an_array[:2, 1:3]  # copy by referrence
a_slice[0,0] = 123
print(a_slice)

[[123  13]
 [ 22  23]]

In [47]:

a_slice = np.array(an_array[:2, 1:3]) #call by value
a_slice[0,0] = 321
print(a_slice)

[[321  13]
 [ 22  23]]

In [8]:

row_rank1 = an_array[1, :]
print(row_rank1, row_rank1.shape)

[21 22 23 24] (4,)

In [9]:

row_rank2 = an_array[1:2, :]    #this gives 2d array
print(row_rank2, row_rank2.shape)

[[21 22 23 24]] (1, 4)

In [11]:

an_array = np.array([[11,12,13], [21,22,23], [31,32,33],[41,42,43] ])

an_array

[[11 12 13]
 [21 22 23]
 [31 32 33]
 [41 42 43]]

In [13]:

col_indices = np.array([0, 1, 2, 0])
print(col_indices)
    
row_indices = np.arange(4)
print(row_indices)

[0 1 2 0]
[0 1 2 3]

In [14]:

#Examine the pairings of row_indices and col_indices.
for row, col in zip( row_indices, col_indices):
    print(row, col)

0 0
1 1
2 2
3 0

In [15]:

# Select one element from each row
an_array[row_indices, col_indices]

Out[15]:

array([11, 22, 33, 41])

In [16]:

an_array[row_indices, col_indices] += 100000
an_array

[[100011     12     13]
 [    21 100022     23]
 [    31     32 100033]
 [100041     42     43]]

In [24]:

an_array = np.array([[11,12], [21,22], [31, 32]])

In [19]:

an_array

Out[19]:

array([[11, 12],
       [21, 22],
       [31, 32]])

In [20]:

# boolean values will be returned as a list
filter = (an_array > 15)
filter

Out[20]:

array([[False, False],
       [ True,  True],
       [ True,  True]])

In [21]:

an_array[filter]

Out[21]:

array([21, 22, 31, 32])

In [22]:

#better way
an_array[an_array > 15]

Out[22]:

array([21, 22, 31, 32])

In [25]:

an_array[an_array > 25] = 25 
an_array

Out[25]:

array([[11, 12],
       [21, 22],
       [25, 25]])

In [26]:

## Data type of an element in array is matter!!
ex1 = np.array([11, 12])
ex1.dtype

Out[26]:

dtype('int64')

In [27]:

ex2 = np.array([11.0, 12.0])
ex2.dtype

Out[27]:

dtype('float64')

In [28]:

ex3 = np.array([11,12], dtype=np.int64)
ex3.dtype

Out[28]:

dtype('int64')

In [29]:

ex4 = np.array([11.1, 12.7], dtype = np.int64)
ex4.dtype

Out[29]:

dtype('int64')

In [31]:

ex5 = np.array([ 11, 12 ], dtype = np.int64)
ex5.dtype
ex5

Out[31]:

array([11, 12])

In [32]:

## ndarray operation

x = np.array([[ 111,112     ],     [121,122 ]], dtype = np.int)
y = np.array([[ 211.1,212.1 ], [221.1,222.1 ]], dtype = np.int)
print(x)
print()
print(y)

[[111 112]
 [121 122]]

[[211 212]
 [221 222]]

In [33]:

x + y

Out[33]:

array([[322, 324],
       [342, 344]])

In [34]:

np.add(x,y)

Out[34]:

array([[322, 324],
       [342, 344]])

In [35]:

np.sqrt(x)

Out[35]:

array([[10.53565375, 10.58300524],
       [11.        , 11.04536102]])

In [36]:

np.exp(x)

Out[36]:

array([[1.60948707e+48, 4.37503945e+48],
       [3.54513118e+52, 9.63666567e+52]])

In [ ]: