Question

Python's Pandas Series/Dataframe Questions (If Statements and Loops are NOT ALLOWED)

Consider the distances.txt (pasted below) file. Such a file can be read into a pandas DataFrame arranged like so:

	Chicago	Los Angeles	New York	Philadephia
Chicago	0	2054	802	738
Los Angeles	2054	0	2786	2706
New York	802	2786	0	100
Philadephia	738	2706	100	0

Write a function named total_distance that accepts a distance matrix (like the one above) and an iterable list of cities. Your function should then calculate the total distance traveled when visiting the cities in that order and returning to the starting city. As with the other exercises, your function should use only vectorized pandas operations and no loops or list comprehensions. Here are some examples:

In [1]: total_distance(dist_matrix,['Chicago','Los Angeles'])

Out[1]: 4108
In [2]: total_distance(dist_matrix,['Chicago','Los Angeles','Philadelphia'])
Out[2]: 5498
In [3]: total_distance(dist_matrix,['Chicago','Los Angeles','Philadelphia','New York'])
Out[3]: 5662
In [4]:total_distance(dist_matrix,['Chicago','Los Angeles','Philadelphia','New York','Los Angeles'])
Out[4]: 9700

The file distances.txt is pasted below:

0,2054,802,738
2054,0,2786,2706
802,2786,0,100
738,2706,100,0.

This problem is often known as calculating a "tour". If you want to have some fun, you can then use your function to solve a famous problem. Can you name it? Here's some sample code:

import pandas as pd
import  itertools
ci t i e s = [ ' Chicago ' , ' Los Angeles ' , 'New York ' , ' Philadelphia ' ]
dist_matrix = pd . read_csv ( ' d i s t a n c e s . txt ' , header=None , names=cities )
di s t_mat r ix . index = c i t i e s
shor t e s t_tour = [ ]
sh o r t e s t_d i s t = 1000000
for a_tour in i t e r t o o l s . permutat ions ( c i t i e s , 4 ) :
di s t anc e = t o t a l_di s t anc e ( di s t_mat r ix , l i s t ( a_tour ) )
i f di s t anc e < sh o r t e s t_d i s t :
sh o r t e s t_d i s t = di s t anc e
shor t e s t_tour = a_tour

print ( sho r t e s t_di s t , shor t e s t_tour )

Point of Stress: This is a Python question requiring the use of Pandas Series or DataFrame (please loops are not allowed - zero score for using loops). Please attach a print screen in addition to the codes to show the scripts ran properly and as intended.

Answer 1

Answer: See the code for function below:

------------------------------------------

import pandas as pd
cities = ['Chicago','Los Angeles','New York','Philadelphia']
dist_matrix = pd.read_csv('distances.txt',header=None,names=cities)
dist_matrix.index = cities

#function to calculate total distance for a given sequence of cities
def total_distance(dist_matrix,cities_list):
    #origin cities
    origins = cities_list
    #print(origins)
    #destination cities
    dests = cities_list[1:len(cities_list)]
    dests.append(cities_list[0])  
    #print(dests)
    #return distance covered
    return sum(dist_matrix.lookup(origins,dests))

tour=['Chicago','Los Angeles']  
print("tour:",tour)
print("distance covered:",total_distance(dist_matrix,tour))
tour=['Chicago','Los Angeles','Philadelphia']  
print("tour:",tour)
print("distance covered:",total_distance(dist_matrix,tour))
tour=['Chicago','Los Angeles','Philadelphia','New York']  
print("tour:",tour)
print("distance covered:",total_distance(dist_matrix,tour))
tour=['Chicago','Los Angeles','Philadelphia','New York','Los Angeles']  
print("tour:",tour)
print("distance covered:",total_distance(dist_matrix,tour))

--------------------------------------------------------------------------------------

Output:

tour: ['Chicago', 'Los Angeles' distance covered: 4108 tour: ["Chicago· 'Los Angeles', distance covered: 5498 philadelphia.] tour: ["Chicago· 'Los Angeles', distance covered: 5662 philadelphia", New York .] tour: ["Chicago· 'Los Angeles", philadelphia", New York', 'Los Angeles'] distance covered: 9700