Pandas Vectorization

Before solving these exercises you should have read High-Level Data Management with Pandas.

import pandas as pd

For these exercises we use the dataset obtained in the Cafeteria project.

data = pd.read_csv('meals.csv', names=['date', 'category', 'name', 'students', 'staff', 'guests'])

data

Preprocessing

Data Types

Convert 'date' and 'category' columns to Timestamp and category, respectively (see Advanced Pandas exercises).

Solution:

# your solution

Count Categories

For each category give the number of meals.

Solution:

# your solution

Remove Categories

Remove Categories which do not contain full-fledged meals descriptions (e.g., 'Salat Bar').

Solution:

# your solution

Remove Allergens and Additives

Most meals contain information on alergenes and additives (numbers in parantheses). Remove the information to get more readably meal descriptions. Implement the removal procedure twice: without and with vectorized string operations. Get and compare execution times.

Solution:

data['name_backup'] = data['name']

%%timeit
# your solution without vectorized string operations

data['name'] = data['name_backup']
data = data.drop(columns=['name_backup'])

%%timeit
# your solution with vectorized string operations

Simplify Meal Descriptions

Create a new column 'simple' from the 'name' column by removing all lower-case words, all punctuation marks and so on. Only words starting with an upper-case letter are allowed.

Solution:

# your solution

All Meals with…

Given a key word (e.g., 'Kartoffel') get the number of meals containing the keyword and print all meal descriptions.

Solution:

# your solution

Meal Plot

For each day get the number of meals containing some keyword (e.g., 'Kartoffel'). Call Series.plot to visualize the result.

Solution:

# your solution