Explore Chicago Bikeshare Data

# coding: utf-8

# Here goes the imports
import csv
import matplotlib.pyplot as plt

# Let's read the data as a list
print("Reading the document...")
with open("chicago.csv", "r") as file_read:
    reader = csv.reader(file_read)
    data_list = list(reader)
print("Ok!")

# Let's check how many rows do we have
print("Number of rows:")
print(len(data_list))

# Printing the first row of data_list to check if it worked.
print("Row 0: ")
print(data_list[0])
# It's the data header, so we can identify the columns.

# Printing the second row of data_list, it should contain some data
print("Row 1: ")
print(data_list[1])

input("Press Enter to continue...")
# TASK 1
# TODO: Print the first 20 rows using a loop to identify the data.
print("\n\nTASK 1: Printing the first 20 samples")
i=0
while (i <=20):
    print(data_list[i])
    i+=1

#Let's change the data_list to remove the header from it.
data_list = data_list[1:]

# We can access the features through index
# E.g. sample[6] to print gender or sample[-2]

input("Press Enter to continue...")
# TASK 2
# TODO: Print the `gender` of the first 20 rows

print("\nTASK 2: Printing the genders of the first 20 samples")
i=0
while (i <=20):
    print(data_list[i][-2])
    i+=1

# Cool! We can get the rows(samples) iterating with a for and the columns(features) by index.
# But it's still hard to get a column in a list. Example: List with all genders

input("Press Enter to continue...")

# TASK 3
# TODO: Create a function to add the columns(features) of a list in another list in the same order
def column_to_list(data, index):
    '''
Function returns a single column from the csv file.
It iterates through lists(data) and gets one peace of information out of each (according to the index)
data= any list
index= int nunmber
    '''
    column_list = []
    for row in data:
        column_list.append(row[index])
    # Tip: You can use a for to iterate over the samples, get the feature by index and append into a list
    # Removing element 0 'Gender'
    return column_list

# Let's check with the genders if it's working (only the first 20)
print("\nTASK 3: Printing the list of genders of the first 20 samples")
print(column_to_list(data_list, -2)[:20])

# ------------ DO NOT CHANGE ANY CODE HERE ------------
assert type(column_to_list(data_list, -2)) is list, "TASK 3: Wrong type returned. It should return a list."
assert len(column_to_list(data_list, -2)) == 1551505, "TASK 3: Wrong lenght returned."
assert column_to_list(data_list, -2)[0] == "" and column_to_list(data_list, -2)[1] == "Male", "TASK 3: The list doesn't match."
# -----------------------------------------------------

input("Press Enter to continue...")

# Now we know how to access the features, let's count how many Males and Females the dataset have
# TASK 4
# TODO: Count each gender. You should not use a function to do that.
male = 0
female = 0
gender_list=column_to_list(data_list, -2)
for gender in gender_list:
    if gender == 'Male':
        male += 1
    elif gender == 'Female':
        female += 1
    else:
        continue

# Checking the result
print("\nTASK 4: Printing how many males and females we found")
print("Male: ", male, "\nFemale: ", female)

# ------------ DO NOT CHANGE ANY CODE HERE ------------
assert male == 935854 and female == 298784, "TASK 4: Count doesn't match."
# -----------------------------------------------------

input("Press Enter to continue...")

# Why don't we creeate a function to do that?
# TASK 5
# TODO: Create a function to count the genders. Return a list
# Should return a list with [count_male, counf_female] (e.g., [10, 15] means 10 Males, 15 Females)
def count_gender(data_list):
#function counts ocurrences of the strings 'Male' and 'Female'
    male = 0
    female = 0
    gender_data_list=column_to_list(data_list, -2)
    for gender in gender_data_list:
        if gender == 'Male':
            male += 1
        elif gender == 'Female':
            female += 1
        else:
            continue
    return [male, female]


print("\nTASK 5: Printing result of count_gender")
print(count_gender(data_list))

# ------------ DO NOT CHANGE ANY CODE HERE ------------
assert type(count_gender(data_list)) is list, "TASK 5: Wrong type returned. It should return a list."
assert len(count_gender(data_list)) == 2, "TASK 5: Wrong lenght returned."
assert count_gender(data_list)[0] == 935854 and count_gender(data_list)[1] == 298784, "TASK 5: Returning wrong result!"
# -----------------------------------------------------

input("Press Enter to continue...")

# Now we can count the users, which gender use it the most?
# TASK 6
# TODO: Create a function to get the most popular gender and print the gender as string.
# We expect to see "Male", "Female" or "Equal" as answer.
def most_popular_gender(data_list):
    #Function uses count_gender function and then compares it results.
    count_gender(data_list)
    if male > female:
        answer = "Male"
    elif female > male:
        answer = "Female"
    else:
        answer = "Equal"
    return answer


print("\nTASK 6: Which one is the most popular gender?")
print("Most popular gender is: ", most_popular_gender(data_list))

# ------------ DO NOT CHANGE ANY CODE HERE ------------
assert type(most_popular_gender(data_list)) is str, "TASK 6: Wrong type returned. It should return a string."
assert most_popular_gender(data_list) == "Male", "TASK 6: Returning wrong result!"
# -----------------------------------------------------

# If it's everything running as expected, check this graph!
gender_list = column_to_list(data_list, -2)
types = ["Male", "Female"]
quantity = count_gender(data_list)
y_pos = list(range(len(types)))
plt.bar(y_pos, quantity)
plt.ylabel('Quantity')
plt.xlabel('Gender')
plt.xticks(y_pos, types)
plt.title('Quantity by Gender')
plt.show(block=True)

input("Press Enter to continue...")

# TASK 7
# TODO: Plot a similar graph for user_types. Make sure the legend is correct.
print("\nTASK 7: Check the chart!")

def count_user_type(data_list):
    #Creates a similar function to count_gender
    subscriber = 0
    customer = 0
    dependent = 0
    gender_data_list=column_to_list(data_list, -2)
    for user_type in user_types_list:
        if user_type == 'Subscriber':
            subscriber += 1
        elif user_type == 'Customer':
            customer += 1
        elif user_type == 'Dependent':
            customer += 1
        else:
            continue
    return [subscriber, customer, dependent]


user_types_list = column_to_list(data_list, -3)
types = ["Subscriber", "Customer",'Dependent']
quantity = count_user_type(data_list)
y_pos = list(range(len(types)))
plt.bar(y_pos, quantity)
plt.ylabel('Quantity')
plt.xlabel('User Type')
plt.xticks(y_pos, types)
plt.title('Quantity by User Type')
plt.show(block=True)

input("Press Enter to continue...")

# TASK 8
# TODO: Answer the following question
male, female = count_gender(data_list)
print("\nTASK 8: Why the following condition is False?")
print("male + female == len(data_list):", male + female == len(data_list))
answer = "Because len(data_list) also comprises the frequency of empty items, that means, when the gender wasn't inserted"
print("Answer:", answer)
print(male)
print(female)

# ------------ DO NOT CHANGE ANY CODE HERE ------------
assert answer != "Type your answer here.", "TASK 8: Write your own answer!"
# -----------------------------------------------------

input("Press Enter to continue...")

# TASK 9
# TODO: Find the Minimum, Maximum, Mean and Median trip duration.
# You should not use ready functions to do that, like max() or min().

# making the list int (all the items were strings)
trip_duration_list = column_to_list(data_list, 2)
i=0
for i in range(len(trip_duration_list)):
    trip_duration_list[i]= int(trip_duration_list[i])


min_trip = 0.
i = 1
min_trip=trip_duration_list[0]
for i in range(len(trip_duration_list)):
    if trip_duration_list[i] < min_trip:
        min_trip = trip_duration_list[i]
    else:
        continue

max_trip = 0.
i = 1
max_trip=trip_duration_list[0]
for i in range(len(trip_duration_list)):
    if trip_duration_list[i] > max_trip:
        max_trip = trip_duration_list[i]
    else:
        continue

mean_trip = 0.
sum_mean_trip=0
for i in range(len(trip_duration_list)):
    sum_mean_trip += trip_duration_list[i]
mean_trip= sum_mean_trip / len(trip_duration_list)

median_trip = 0.
median_trip_list= sorted(trip_duration_list)
if len(median_trip_list) % 2 == 0:
    median_trip= (median_trip_list[len(median_trip_list)/2] + median_trip_list[len(median_trip_list)/2+ 1]) /2
else:
    median_trip= median_trip_list[round(len(median_trip_list)/2)]


print("\nTASK 9: Printing the min, max, mean and median")
print("Min: ", min_trip, "Max: ", max_trip, "Mean: ", mean_trip, "Median: ", median_trip)

# ------------ DO NOT CHANGE ANY CODE HERE ------------
assert round(min_trip) == 60, "TASK 9: min_trip with wrong result!"
assert round(max_trip) == 86338, "TASK 9: max_trip with wrong result!"
assert round(mean_trip) == 940, "TASK 9: mean_trip with wrong result!"
assert round(median_trip) == 670, "TASK 9: median_trip with wrong result!"
# -----------------------------------------------------

input("Press Enter to continue...")

# TASK 10
# Gender is easy because usually only have a few options. How about start_stations? How many options does it have?
# TODO: Check types how many start_stations do we have using set()
user_types = set(column_to_list(data_list, 3))

print("\nTASK 10: Printing start stations:")
print(len(user_types))
print(user_types)

# ------------ DO NOT CHANGE ANY CODE HERE ------------
assert len(user_types) == 582, "TASK 10: Wrong len of start stations."
# -----------------------------------------------------
input("Press Enter to continue...")
# TASK 11
# Go back and make sure you documented your functions. Explain the input, output and what it do. Example:
# def new_function(param1: int, param2: str) -> list:
'''
      Example function with annotations.
      Args:
          param1: The first parameter.
          param2: The second parameter.
      Returns:
          List of X values

'''

# TASK 12 - Challenge! (Optional)
# TODO: Create a function to count user types without hardcoding the types
# so we can use this function with a different kind of data.
print("Will you face it?")
answer = "yes"

i=0
def count_items(column_list):
    '''
unique_items takes the single elements of column_list with no repetition
the first 'for' adds the elements of unique_items to item_types
the second 'for' adds the number of occurrences of each element of item_types in the column list
to count_items.

    '''
    unique_items= set(column_list)
    item_types = []
    count_items = []
    for unique_item in unique_items:
        item_types.append(unique_item)
    for i in range(len(item_types)):
        count_items.append(column_list.count(item_types[i]))
    return item_types, count_items



if answer == "yes":
    # ------------ DO NOT CHANGE ANY CODE HERE ------------
    column_list = column_to_list(data_list, -2)
    types, counts = count_items(column_list)
    print("\nTASK 11: Printing results for count_items()")
    print("Types:", types, "Counts:", counts)
    assert len(types) == 3, "TASK 11: There are 3 types of gender!"
    assert sum(counts) == 1551505, "TASK 11: Returning wrong result!"