In [1]:

    

##Some code to run at the beginning of the file, to be able to show images in the notebook
##Don't worry about this cell

#Print the plots in this screen
%matplotlib inline 

#Be able to plot images saved in the hard drive
from IPython.display import Image 

#Make the notebook wider
from IPython.core.display import display, HTML 
display(HTML("<style>.container { width:90% !important; }</style>"))


    

In [2]:

    

##this is a list
print([1,2,3])


    

    




[1, 2, 3]

In [2]:

    

print(type([1,2,3]))


    

    




<class 'list'>

In [3]:

    

# A list can combine several data types
this_is_list1 = [3.5,"I'm another string",4]
print(this_is_list1)

# It can even combine several data structures, for instance a list inside a list
this_is_list2 = [3.5,"I'm another string",4,this_is_list1]
print(this_is_list2)


    

    




[3.5, "I'm another string", 4]
[3.5, "I'm another string", 4, [3.5, "I'm another string", 4]]

In [1]:

    

## A list
this_is_a_list = [1,3,2,"b"]
print("Original: ", this_is_a_list)


    

    




Original:  [1, 3, 2, 'b']

In [2]:

    

## Add elements
this_is_a_list.append("c")
print("Added c: ", this_is_a_list)


    

    




Added c:  [1, 3, 2, 'b', 'c']

In [11]:

    

## Get element. The first element has index 0, which means that this_is_a_list[0] gets the first element
print("Fourth element: ", this_is_a_list[3])


    

    




Fourth element:  b

In [7]:

    

this_is_a_list


    

    
Out[7]:





[0, 1, 2, 3, 4]

In [6]:

    

this_is_a_list[1:3]


    

    
Out[6]:





[1, 2]

In [ ]:

    




    

In [4]:

    

#All list
this_is_a_list = [0,1,2,3,4]
print(this_is_a_list)
#"Second to end element (included)"
print("Second to end element: ", this_is_a_list[1:])


    

    




[0, 1, 2, 3, 4]
Second to end element:  [1, 2, 3, 4]

In [5]:

    

#Second to the fourth (included)
print("Second to the fourth (included): ",this_is_a_list[1:4])

#First to the last element (not included)
print("First to the last element (not included): ",this_is_a_list[:-1])


    

    




Second to the fourth (included):  [1, 2, 3]
First to the last element (not included):  [0, 1, 2, 3]

In [6]:

    

print("Original: ", this_is_a_list)
## Remove 4th element and save it as removed_element
removed_element = this_is_a_list.pop(3)
print(removed_element)
print("The list is now: ", this_is_a_list)


    

    




Original:  [0, 1, 2, 3, 4]
3
The list is now:  [0, 1, 2, 4]

In [8]:

    

this_is_a_list = [1,2,3,4,5]


    

In [9]:

    

#Search
print(3 in this_is_a_list)


    

    




True

In [10]:

    

#Find index
print(this_is_a_list.index(4))


    

    




3

In [10]:

    

## Count the number of elements in a list
this_is_a_list = [1, 3, 2]
len_this_is_a_list = len(this_is_a_list) #you tell the computer to sum it and save it as `sum_this_is_a_list` 
print("Length: ", len_this_is_a_list)


    

    




Length:  3

In [17]:

    

## Sort a list
this_is_a_list = [1, 3, 2]
this_is_a_list = sorted(this_is_a_list) #you tell the computer to sort it, and to save it with the same name
print("Sorted: ", this_is_a_list)


    

    




Sorted:  [1, 2, 3]

In [16]:

    

## Sum a list
this_is_a_list = [1, 3, 2]
sum_this_is_a_list = sum(this_is_a_list) #you tell the computer to sum it and save it as `sum_this_is_a_list` 
print("Sum: ", sum_this_is_a_list)


    

    




Sum:  6

In [6]:

    

sum(["1","2"])


    

    




---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-6-e54effb7b228> in <module>()
----> 1 sum(["1","2"])

TypeError: unsupported operand type(s) for +: 'int' and 'str'

In [11]:

    

print(float("1"))


    

    




1.0

In [29]:

    

## Create a list [0,1,2,3,4]
print(list(range(0,5,1)))
print(range(5)) #for all practical issues you don't need to convert them


    

    




[0, 1, 2, 3, 4]
range(0, 5)

In [12]:

    

## Create the list [1,2,3,4]
print(list(range(1,5,1)))


    

    




[1, 2, 3, 4]

In [13]:

    

## Create the list [1,3,4,5,7,9]
print(list(range(1,10,2)))


    

    




[1, 3, 5, 7, 9]

In [4]:

    

import numpy as np
np.arange(0,5,0.5)


    

    
Out[4]:





array([ 0. ,  0.5,  1. ,  1.5,  2. ,  2.5,  3. ,  3.5,  4. ,  4.5])

In [278]:

    

#we create a string and call it "eggs and bacon"
our_string = "eggs and bacon"
#now we divide it into words, creating a list and saving it with the name our_list_of_wrods
our_list_of_words = our_string.split()
print(our_list_of_words)


    

    




['eggs', 'and', 'bacon']

In [279]:

    

#we can do the opossite and join the words using the function "join". 
" ".join(our_list_of_words)


    

    
Out[279]:





'eggs and bacon'

In [280]:

    

#we can join the words using any character we want in the quoted par of (" ".join)
":::".join(our_list_of_words)


    

    
Out[280]:





'eggs:::and:::bacon'

In [23]:

    

#we can also divide a string using other characters instead of space. 
#for example let's split in the "a"s
our_string = "eggs and bacon"
print(our_string.split("a"))


    

    




['eggs ', 'nd b', 'con']

In [15]:

    

#we can change parts of the word
our_string = "eggs and bacon"
print(our_string.replace("bacon","tofu"))


    

    




eggs and tofu

In [286]:

    

#we can find where a word start. For example let's find how many characters there are before "and"
our_string = "eggs and bacon and"
print(our_string.find("and"))


    

    




5

In [16]:

    

#and we can use this to slice the string like we did with the list
our_string1 = "10 events"
our_string2 = "120 events"
our_string3 = "2 events"

#we find the index of and
index_and1 = our_string1.find("event")
index_and2 = our_string2.find("event")
index_and3 = our_string3.find("event")

print(index_and1)
print(index_and2)
print(index_and3)


    

    




3
4
2

In [17]:

    

print(our_string1)


    

    




10 events

In [21]:

    

#and keep all the string until that index
print(our_string1[:our_string1.find("event")])
print(our_string2[:index_and2])
print(our_string3[:index_and3])


    

    




10 
120 
2 

In [18]:

    

this_is_a_list?


    

In [20]:

    

our_string1.find?


    

In [30]:

    

this_is_a_tuple = (1,3,2,"b")
print(this_is_a_tuple)
this_is_a_list = list(this_is_a_tuple)
print(this_is_a_list)


    

    




(1, 3, 2, 'b')
[1, 3, 2, 'b']

In [33]:

    

#If we try to pop and elment, like with lists, we get an error.
this_is_a_tuple.pop(0)


    

    




---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-33-81256b9b9870> in <module>()
      1 #If we try to pop and elment, like with lists, we get an error.
----> 2 this_is_a_tuple.pop(0)

AttributeError: 'tuple' object has no attribute 'pop'

In [272]:

    

Image(filename='./images/set-operations-illustrated-with-venn-diagrams.png')


    

    
Out[272]:

In [12]:

    

{1,2,3} - {2,5,6}


    

    
Out[12]:





{1, 3}

In [273]:

    

#Let's create a list with repeated elements
this_is_a_list = [3,1,1,1,2,3]
print(this_is_a_list)
#Now we convert it to a set and see how the repeated elements are gone
this_is_a_set1 = set(this_is_a_list)
print(this_is_a_set1)
#You can also create sets like this
this_is_a_set2 = {1,2,4}
print(this_is_a_set2)


    

    




[3, 1, 1, 1, 2, 3]
{1, 2, 3}
{1, 2, 4}

In [183]:

    

## Union
print(this_is_a_set1 | this_is_a_set2)


    

    




{1, 2, 3, 4}

In [184]:

    

## Intersection
print(this_is_a_set1 & this_is_a_set2)


    

    




{1, 2}

In [185]:

    

## Diference set_1 - set2
print(this_is_a_set1 - this_is_a_set2)


    

    




{3}

In [275]:

    

## Diference set_2 - set1
print(this_is_a_set2 - this_is_a_set1)


    

    




{4}

In [276]:

    

## Very useful for words. Imagine we have two articles, one saying "eggs bacon python" and another saying "bacon spam" 
#We can find which words they have in common
print({"eggs","bacon","python"} & {"bacon","spam"})


    

    




{'bacon'}

In [14]:

    

#First we need to import it
import numpy as np


    

In [15]:

    

#Sum of a list
a_list = [0,1,2,3,4,5,6]
#Using the standard function
print(sum(a_list))
#Using numpy
print(np.sum(a_list))


    

    




21
21

In [16]:

    

##How to find the mean() of a list of numbers? mean() does not exist
print(mean(a_list))


    

    




---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-16-37fa84e18150> in <module>()
      1 ##How to find the mean() of a list of numbers? mean() does not exist
----> 2 print(mean(a_list))

NameError: name 'mean' is not defined

In [23]:

    

a_list = [0,1,2,3,4,5,6]
a_np_array = np.array(a_list)
a_np_array**10


    

    
Out[23]:





array([       0,        1,     1024,    59049,  1048576,  9765625, 60466176])

In [16]:

    

##but numpy rescues you
import numpy as np

a_list = [0,1,2,3,4,5,6]

#first convert the list to an array. 
#This is not required when you use np.mean(a_list), but it is required in some other ocassions. 
a_np_array = np.array(a_list) 

print(type(a_list))
print(type(a_np_array))
print(np.mean(a_np_array))


    

    




<class 'list'>
<class 'numpy.ndarray'>
3.0

In [17]:

    

##you can take powers
print(a_np_array**2) #this would not work with a list


    

    




[ 0  1  4  9 16 25 36]

In [257]:

    

##or square roots
print(np.sqrt(a_np_array)) #this would work with a list


    

    




[ 0.          1.          1.41421356  1.73205081  2.          2.23606798
  2.44948974]

In [252]:

    

#or some basic statistics
import numpy as np
import scipy.stats #another library for more complicated statistics

#we create a list with numbers 0,1,2...998,999
numElements = 1000
this_is_an_list = list(range(numElements)) 
this_is_an_array = np.array(this_is_an_list)
print(this_is_an_array[:10]) #print only the 10 first elements to make sure it's okay

#and print some stats
print(np.mean(this_is_an_array))
print(np.std(this_is_an_array))
print(np.median(this_is_an_array))
print(scipy.stats.mode(this_is_an_array))
print(scipy.stats.skew(this_is_an_array))


    

    




[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
499.5
288.674990257
499.5
ModeResult(mode=array([0]), count=array([1]))
0.0

In [24]:

    

#Let's start with this array
this_is_an_array = np.array([1,2,3,4,5,6,7,8,9,10])
print(this_is_an_array)


    

    




[ 1  2  3  4  5  6  7  8  9 10]

In [109]:

    

#If we want the elements greater or equal (>=) to 5, we could do:
print(this_is_an_array[4:])


    

    




[ 5  6  7  8  9 10]

In [25]:

    

#However this case was very easy, but what getting the elements >= 5 in here:
#[ 5,  9,  4,  3,  2,  8,  6, 7, 10,  1]
unsorted_array = np.array([ 5,  9,  4,  3,  2,  8,  6, 7, 10,  1])
print(unsorted_array)


    

    




[ 5  9  4  3  2  8  6  7 10  1]

In [26]:

    

unsorted_array >= 5


    

    
Out[26]:





array([ True,  True, False, False, False,  True,  True,  True,  True, False], dtype=bool)

In [30]:

    

unsorted_array[np.array([ True,  True, False, False, False,  True,  True,  True,  True, False])]


    

    
Out[30]:





array([ 5,  9,  8,  6,  7, 10])

In [261]:

    

#We can do the following: unsorted_array[unsorted_array >= 5]
#which means keep the elements of unsorted_array that are larger or equal to 5
#unsorted_array[condition]
print(unsorted_array[unsorted_array >= 5])


    

    




[ 5  9  8  6  7 10]

In [116]:

    

#This is a special kind of slicing where you filter elements with a condition. 
#Lists do not allow you to do this
#How does it work? By creating another array, of the same lenght, with False and Trues
print(unsorted_array)
print(this_is_an_array >= 5) #the same than comparing 3 >= 5, but numpy compares every number inside the array.


    

    




[ 5  9  4  3  2  6  7  8 10  1]
[ True  True False False False  True  True  True  True False]

In [120]:

    

#We can use variables
condition = this_is_an_array >= 5 #the computer does the = at the end always
print(unsorted_array[condition])


    

    




[ 5  9  6  7  8 10]

In [125]:

    

#What if we want to get the numbers between 5 and 9 (6, 7,9)?
unsorted_array = np.array([ 5,  9,  4,  3,  2,  8,  6, 7, 10,  1])

#We do it with two steps

#Step 1: Greater than 5
condition_gt5 = unsorted_array > 5
unsorted_array_gt5 = unsorted_array[condition_gt5]
print("Greater than 5", unsorted_array_gt5)

#Step 2: Lower than 9
condition_lw9 = unsorted_array_gt5 < 9 #we are using the new array
unsorted_array_gt5_lw9 = unsorted_array_gt5[condition_lw9]
print("Greater than 5 and lower than 9", unsorted_array_gt5_lw9)


    

    




Greater than 5 [ 9  6  7  8 10]
Greater than 5 and lower than 9 [6 7 8]

In [25]:

    

unsorted_array = np.array([ 5,  9,  4,  3,  2,  8,  6, 7, 10,  1])
condition_gt5 = unsorted_array > 5
condition_lw9 = unsorted_array < 9

print(unsorted_array[condition_gt5 & condition_lw9])


    

    




[8 6 7]

In [31]:

    

##first we import it
import pandas as pd


    

In [19]:

    

Image(url="http://www.relatably.com/m/img/boring-memes/when-my-friend-tell-me-a-boring-story_o_1588863.jpg")


    

    
Out[19]: