Install Python from Python.org, choose an installer based on your OS.
We use PyCharm in the course. It's a great and free source code editor for Python that makes your life easier. Similar to a word processor for text it has integrated spell checking and can help you solve programming errors while you type.
A Python script is interpreted from top to bottom. The Python interpreter reads your source code and translates it on the fly into machine code that your computer understands. There is no compilation step, since the code is interpreted at runtime.
Open PyCharm, make a new project (uncheck the checkbox that says something about main.py). Now add a new Python file by right clicking in the project tree on the left side and chose new Python file. You can give it any name you want, by convention the Python script that should be executed in a project is often named main.py .
greeting = 'hello world'
greeting = 'how are you?'
print(greeting)store simple values in memory:
count = 7
rating = 4.9
name = 'Benjamin'
is_published = True #or
is_published = False True / False: boolean values. Case sensisive! true / false will give an error. The 4 most basic types in Python are:
Integers (int): 42
Floating numbers (float): 4.67869
Strings (str): 'test'
Boolean (bool): True or False
They are used depending on the information that you need to store in memory. Its also possible to store more complex objects in variables and we'll discuss this in the future.
Excercise:
We check in a patient named Alan Smithee. He's 30 years old and is a new patient.
name = 'Alan Smithee'
# or
name = "Alan Smithee"
age = 30
is_new_patient = TrueStrings store text in Python, they are declared with either ' or ". The text is placed inside '...' or "...".
print('Hello World!')The print functions and others are always available, they are built in functions of the python interpreter.
The input function can be used to request user input on the command line:
input('What is your name? ') # the () <- are executing the function. Parameters are inside the () and can be seperated by a , This program asks the users name and says hi !
name = input('What is your name? ')
print('Hi ' + name + '!')Strings can be concatenated with +
birth_year = input('Birth year:')
age = 2021 - birth_year
print('Your age is:' + age)--> this will results in a type_error
Because internally, Python will store the birth year as:
birth_year = '1987'A variable can be of any type. Not all operations work on all types, in particular combining two different types may be ambigious. To solve this issue we can use builtin functions that concert a particular type (type conversion) and explicity tell Python how to handle a calculation:
int('1987') # -> 1987There are more type functions:
int('7') # -> 7
float('5.9') # -> 5.9
str(5.9) # -> "5.9"You can check that the type of birth year is string, by using the type function:
print(type(birth_year))The complete program, that calculates the age correctly with type conversions:
birth_year = input('Birth year:')
age = 2021 - int(birth_year)
print('Your age is:' + str(age))course = 'Python for beginners'
course = 'Python's ...' #-> error, python doesn
course = "Python's ..." #-> correct way to store a string containing a '
course = 'Python\'s ...' #-> alternative, use \ to backspace the 'Triple quotes can be used for very long lines of text:
text = '''Very long line of text
that can also span
multiple lines'''course = ‘Python for beginners’
01234The slicing operator [] can be used to 'slice' individual characters from a string, or slice a sequence of characters from a string. Each character in a string has an index, starting from 0.
print(course[0])-> 'P'
print(course[-1])Slice the last character with -1 (counts backwards from the end). -> 's'
print(course[0:3])Use [x:y] to select a range of characters from x to y:
-> 'Pyt'
print(course[1:])Or [x:] to select from x till the end.
print(course[:5])Or [:y] to select from the beginning to y.
first = 'Alan'
last = 'Smithee'
address = 'ABC street 1'
msg = 'Hi ' + first + ' ' + last + ', ' + 'your package has been sent to: ' + address
print(msg)String formatting is more convienent, when you have to combine many strings. You can copy content from variables into the string using {}:
msg = f'hi {first} {last}, your package has been sent to: {address}'
print(msg)course = 'Python for beginners'Returns the length of a string:
len(course)Try these methods:
.find("beginners")
.upper()
.lower()
functions vs. methods: strings and other object have methods that are particular to the the object. For example a string has methods that over objects don't have in Python. Such methods are called with the variable name (or string directly), follow by a . and the method name, then (). Or (arg1), (arg1, arg2) if there are any arguments. See these examples:
course = 'Python for beginners'
print(course.upper())
print(course.lower())
print(course.title())
print(course.find('o'))
print(course.find('O'))
print(course.find('beginners'))The replace function is useful to replace any occurence of a string with another:
.replace('beginners', 'absolute beginners')if statements can be used to conditionally execute another block of python code. The condtional code is intended by one level. The geneneral synytax is:
if <condition>:
<conditional block of code>
...
elif <condition2>:
<conditional block of code>
...
elif <condition3>:
<conditional block of code>
...
else:
<conditional block of code, if none of the above was True>There can be multiple optional else if (elif) statements. They will only execute the conditioal code if none of the previous condition were met and the current condition is true. Once one if or elif executes the conditional block, none of the other statements will be checked. There can only be one optional else statement, that will exceute if none of the if or elif conditions was met.
Code example, depending on the temperature, print a different message to the user:
if celsius < 0:
print("It's freezing!!! :( You need a jacket. A warm one.")
elif celsius < 22.0:
print(':(')
else:
print('You only need a t-shirt')
print(':)')If statements - excercise: convert
Excercise: ask the user
Solution:
command = input('Convert meter to feet (mtf)'
' or feet to meters (ftm)?'
' Please answer: ftm or mtf! ')
if command == 'ftm':
feet = input('Please input feet:')
meter = float(feet) / 3.28408
print('result in meters is:', meter)
elif command == 'mtf':
meter = input('Please input meters:')
feet = float(meter) * 3.28408
print('result in feet is:', feet)
else:
print('Not a valid command!')Example: Test if word is contained in string. This example also uses the in operator, to test if a string is contained in another string.
course = 'Python for beginners'
if 'beginners' in course:
print('This is a beginners course')
elif 'advanced' in course:
print('This is an advanced course')i = 1
while i <=5:
print(i) -> print('*' * i)
i = i+1In a while loop, all intended statements are looped until the condition at the top returns False (= is no longer valid / True). You can read it as: do this while .
break can be used to abort a loop. continue can be used to break the current flow through the loop to restart the next iteration at the top of the loop.
A while loop can use an else: part as well, that is executed if no break occured.
exercise loops #1 Guessing game exercise: (while loop with break statement)
Solution:
import random
random_num = random.random()
random_num *= 10.0
secret = int(random_num)
# keep track if the user won or lost
has_lost = True
# Give the user 3 tries to guess the secret
while counter < 3:
counter += 1
secret_test = input("What is the secret number (0-9)? ")
if secret == int(secret_test):
print("Wow yes thats true! You have won")
has_lost = False
break
print("Remaining tries:", 3-counter)
if has_lost:
print("Guessed wrong 3 times, you lost.")Import statements: additional functionality can be loaded with the import statement. General syntax is import . For example, to generate and print a random number between 0.0 (inclusive) and 1.0 (exclusive):
import random
print(random.random())exercise loops #2, simulate a car.
Another very frequent loop in Python is the for loop. It is usually a more convient way to describe a loop over a sequence, try the following code:
for num in [0, 1, 2, 3, 4]:
print(num)In the above example, a list is used to describe the sequence that is iterated on. In each run through the for loop, num changes according to the items in the list. You can store sequences of any Python types/objects in a list. You'll often encounter list with basic types such as intergers, strings or float. For example [0, 1, 2, 3, 4] is a list of 5 integer numbers.
Since iterating over a sequence of incrementing integers is a very common thing to do, Python has the short hand function range(...):
range(5) # iterator that iterates over [0, 1, 2, 3, 4]
range(5, 10) # iterator that iterates over [5, 6, 7, 8, 9]
range(5, 10, 2) # iterator that iterates over [5, 7, 9]This means, to print the numbers
for num in range(5):
print(num)And from 0 to 99:
for num in range(100):
print(num)WIP the following text is work in progress and not yet completed WIP
excerise prices: sum prices with a for loop:
prices = [4.99, 12.0, 0.79, 19.99]
sum = 0.0
for ...
print("sum:",sum)loops: nested for loops
numbers = [5, 2, 5, 2, 2]
for x_count in numbers:
print('*'*x_count)
output = ''excercise: replace '*'*x_count with a for loop
In the previous excercise, we encountered a list of floats. Lists can have elements of any type in Python. In particular and very relevant to this course, a list of strings can also be defined, for example:
names = ['Sofia', 'Bob' , 'Eva', 'Adam' ]This defines a list of four names. Try these:
print(names)
print(names[2])
print(names[2:])Similarly to slicing in a string, the [] operator can also be applied to lists. Indexing starts at 0 and counts the elements of the list:
names = ['Sofia', 'Bob' , 'Eva', 'Adam' ]
0 1 2 3Lists can also be nested, i.e. a list can also contain other lists:
names = [['Sofia', 'Eva'], ['Bob', 'Adam']]Try these:
print(names[0])
print(names[1])Items can be inserted, removed and appended to a list:
names = ['Sofia', 'Bob' , 'Eva', 'Adam' ]
names.insert(0, 'Nina')
print(names)
names.remove(0)
print(names)
names.append('John')
print(names)Try these list methods, what do they do?
.pop()
.clear()
.index("Bob")
The in operator can also be used on list items, to check if a particular element is contained in a list, e.g.
if 'Bob' in names:
print('Bob is in the list of names!')Lists can also be sorted, reversed and copied:
names.sort()
print(names)
names.reverse()
print(names)
names_copy = names.copy()
print(names_copy)A string can also be split by white space into a list, try this:
test = 'This is a test'.split()
print(test)Closer look: lists exercise
Write a program to remove repeating entries from this list:
numbers = [4,5,5,3,3,4]Tuples are like lists, but the elements can't be changed after the tuple was created. Tuples are immutable in Python, while lists are mutable. A tuple is created with () instead of [].
coordinates = (4, 3)x = coordinates[0]
y = coordinates[1]
x, y = coordinates
print(x*y)Dictionaries store key/value pairs in Python. Searching for a key in a dictionary is very efficient in Python and fast even if a dictionary is really big. For example, to create a phonetic dictionary in Python, the keys would be words and the values the IPA sequence for a word. Another example would be telephone book, where the names are
telephone_book = {'Adam':'+49 48264869', 'Eva':'+49 4738384', 'Bob':'+49 685685685', }Dictionaties are initialized with {} containing a list of key:value pairs. Additional entries can also be added later on:
telephone_book['Alice'] = '+49 567373577'To retrieve the value for the key 'Bob', you can do:
name = 'Bob' print(telephone_book[name])
#or
print(telephone_book['Bob'])
The in operator can be used to check if a key is in a dictionary:
if 'Bob' in telephone_book:
print('Bob is in the telephone book!', telephone_book['Bob'])
else:
print('Bob is not in the telephone book :(', telephone_book['Bob'])A dictionary can also be empty at first, with key/values added later on:
empty_dictionary = {}Write a digit converter, that translates a sequence of digits into words, for example: 02585 -> "zero two five eight five"
def unique(input_list):
new_list = []
for item in input_list:
if item not in new_list:
new_list.append(item)
return new_list
print(unique([5,5,5,42,5,3,42]))def add(num1, num2):
return num1+num2
print(add(4,6))The def statement can be used to declare your own functions. This encapsulates functionality, that you can later reuse.
Syntax is:
def functioname():or
def functioname(argument1):or
def functioname(argument1, argument2):Any code inside the function is indented one level. The return keyword is used to end the function call and return one or more objects.
return elemAfter your function is declared, you can use your function by calling it with the name you have given it, followed by () and potential arguments:
functioname(argument1, argument2)Write a smiley converter, that converts :) and :(, :D in a text into emoticons: 😞 😃 😛
This is similar to dicts exercise 1. However, this time write a new function that takes a string as input and produces the new string with the replacements as output. Then call your function to test it and print some example results.
In python, we use the following syntax to open a text file and read it line by line:
with open('names.txt') as names:
for line in names:
print(line)The open() function is builtin, similar to print input etc. The second argument to open is by default 'r' for read, so the above is identical to:
with open('names.txt', 'r') as names:
for line in names:
print(line)Text files can also be written to:
with open('numbers.txt', 'w') as numbers_out:
for number in range(10):
numbers_out.write(str(number) + '\n') # The '\n' character makes a new line. While print automatically appends a new line, .write() does not do this.with open('names.txt') as names, open('names_upper.txt','w') as names_out:
for line in names:
names_out.write(line.upper())Load the names dataset from http://shapecatcher.com/names.zip
We want to analyzse the distribution of characters at the end of English female and male names.
You can start from this template, that reads the two files female.txt and male.txt line by line (and just prints the content):
def analyse(names_file):
with open(names_file) as names_female:
for line in names_female:
if line[-1] == '\n':
line = line[:-1]
name = line
print(name)
analyse("names/female.txt")
analyse("names/male.txt")def get_second_element(item):
return item[1]
def analyse(names_file):
corp = {}
with open(names_file) as names:
for line in names:
line = line.strip() # -> strips \n
last_character = line[-1]
if last_character not in corp:
corp[last_character] = 1
else:
corp[last_character] += 1
corp_sorted = dict(sorted(corp.items(),
reverse=True,
key=get_second_element))
print(corp_sorted)Default dicts are like regular dictionaries, but every key that is not available is automatically generated and initialized (to zero for int):
from collections import defaultdict
dd = defaultdict(int)
print(dd["test"])from collections import defaultdict
def get_second_element(item):
return item[1]
def analyse(names_file):
corp = defaultdict(int)
with open(names_file) as names:
for line in names:
line = line.strip() # -> strips \n
last_character = line[-1]
corp[last_character] += 1
corp_sorted = dict(sorted(corp.items(),
reverse=True,
key=get_second_element))
print(corp_sorted)
print('female names:')
analyse("names/female.txt")
print('male names:')
analyse("names/male.txt")Load the shakespeare dataset from http://shapecatcher.com/shakespeare.zip
We will work with Spacy, a module for various NLP tasks:
import spacy
spacy.cli.download("en_core_web_sm")
nlp = spacy.load("en_core_web_sm")
with open("shakespeare.txt") as input_file:
text = input_file.read()
shakespeare_corpus = nlp(text)
for token in shakespeare_corpus:
print(token)Excercise: What are the most frequent words in Shakespears complete works?
Install matplotlib first, you should be able to write:
import matplotlib In PyCharm and it should automatically suggest to install matplotlib. In this course we'll use a subpackage of matplotlib called pyplot, that provides an easy interface to the plotting functionailty in matplotlib.
Examples to create line plots:
import matplotlib.pyplot as plt
x_data = [4., 7., 9., 15.7]
y_data = [x*x for x in x_data]
x_data2 = [x/2.5 for x in x_data]
plt.title("My first plot")
plt.plot(x_data, y_data, 'o-', color='green')
plt.plot(x_data2, y_data, 'x-', color='red')
plt.xlabel("number of tries")
plt.ylabel("seconds")
plt.legend(["Experiment 1","Experiment 2"], loc="lower right") Histogram plot example using 1000 artifically generated datapoints, normally distributed:
import numpy as np
import matplotlib.pyplot as plt
# generate a random list with normally distributed datapoints, mean of 100 and sigma of 15
mu = 100.
sigma = 15.
x = mu + sigma * np.random.randn(1000)
print(x) # random numbers centered around 100
# plot a histogram with 100 bars
plt.hist(x, 100)
plt.show()Example plotting frequency/rank of words in a vocabulary dictionary:
# this would usually be generated from real data, needs to be sorted!
vocabulary = {'the':523, 'cat':235, 'dog':132}
# generate a list with 1,2,3 for the ranks in the x axis
x_data = list(range(1,4))
# retrieve the frequency of words from the dictionary
y_data = [vocabulary[word] for word in vocabulary]
# check that data looks correct:
print(x_data)
print(y_data)
# plot it as a green line (g) with oval data markers (o) and a line connecting the data points (-)
plt.plot(x_data, y_data, 'go-')
# manually set limits on the axis:
plt.xlim(1,10)
# show the plot the user
plt.show()We will start with a new corpus, a dataset of movie reviews from IMDB. Download it here https://shapecatcher.com/IMDBdata.zip
This dataset contains reviews + sentiment (positive / negative). We want to find out which words (adjectives) are frequently used in positive reviews and which ones are used in negative reviews. We would also like to see if there are different patterns in how long (number of words) a negative or positive reviews is and would like to plot the lengths in a histogram plot to learn more about this.
This new dataset is in csv format. Csv stand for comma separated values and is a common format for tabular data. There is also a nice Python module to read csv files:
import csv
with open('IMDB Dataset.csv', newline='') as csvfile:
reader = csv.reader(csvfile)
for row in reader:
print('text:',row[0])
print('sentiment:',row[1]) This, we try to structure our code into functions:
import csv
def process_csv():
with open('IMDB Dataset.csv', newline='') as csvfile:
reader = csv.reader(csvfile)
for row in reader:
print('text:',row[0])
print('sentiment:',row[1])
if __name__ == "__main__":
process_csv()For the excercise:
import csv
def analyze_adjectives(dataset):
#YOUR CODE HERE
return
def plot_length_distribution(dataset):
#YOUR CODE HERE
return
def read_dataset():
dataset = []
with open('IMDB Dataset.csv', 'r', newline='', encoding='utf-8') as csvfile:
reader = csv.reader(csvfile)
for row in reader:
review = row[0]
sentiment = row[1]
dataset.append((review, sentiment))
return dataset
if __name__ == "__main__":
print('Read dataset')
imdb_review = read_dataset()
print('Done')
analyze_adjectives(imdb_review)
plot_length_distribution(imdb_review)Typically a larger Python program has multiple functions, potentially distributed accross many files. The main file will have a statement:
if __name__ == "__main__":
#main entry pointat the end, this is the main entry point of the program.
Kokensparger, Brian. Guide to Programming for the Digital Humanities: Lessons for Introductory Python. Cham: Springer International Publishing, 2018.
https://doi.org/10.1007/978-3-319-99115-3
http://emedien.sub.uni-hamburg.de/han/SpringerEbooks/doi.org/10.1007/978-3-319-99115-3
Hammond, Michael. Python for Linguists. Cambridge University Press.
https://doi.org/10.1017/9781108642408
http://emedien.sub.uni-hamburg.de/han/CambridgeBooksOnline/doi.org/10.1017/9781108642408