Python for Linguists

Students are introduced to python, and the different techniques to process natural language texts automatically are explained. Experience in programming is desirable, but not necessary. Since the course is open to students without prior programming experience, sessions 2-4 are very general and not specific to natural language text processing.

There will be 12 exercises and 13 explanation sessions, which can be found on the course repository on github.com. The first session is dedicated to the explanation of git in general and how we are using it in the course: All the exercises are handed in via pull requests and will be tested automatically. If a submission passes the tests, it is considered a success; and for the maximum grade students need to submit 10 successful exercises. Since we are using pytest to automatically evaluate your submissions, avoid naming any of your files beginning with test_ or ending with _test.

Table of Contents

• exercise-1 Introduction to git

• exercise-2 Introduction and installation of python

• exercise-3 Data types: string and list

• exercise-4 dict

• exercise-5 comprehensions, reading and writing files, .csv

• exercise-6 import and arguments

• exercise-7 Good practises (PEP8)

• exercise-8 virtual environments:pip and poetry

• exercise-9 jupyter

• exercise-10 matplotlib for visualisation

• exercise-11 nltk and sentiment analysis

• exercise-12 spaCy and dependency parsing

• exercise-13 word embeddings

How to use git for this course

To get the new exercise, you have to create a pull request from the course repository into you own repository. Then, pull the new files into your repl.

I recommend that the students use the web interface of github.com and replit.com to manage their commits, so that they don't have to deal too much with the intricacies of git and can focus on the programming.

Should there be problems with the commit history and with pull requests, check the following points:

• Make sure you're on your fork of the project and that the pull request runs in the right direction (YourName / uibk-python ⬅️ Aequivinius / uibk-python for getting the new exercise, and the other way for submitting your solution).
• If you receive a message ❌ Can't automatically merge, issue the pull request anyway.

• You'll receive a list of conflicting files, which you delete.

• Then you should be able to then proceed with your pull request. Should the problem persist, check the Files changed tab, ⠐⠐⠐ and View file for the file that is causing the conflict. Copy its contents and paste them in your repository at the correct location.

• As a method of last resort you can always delete your repository (⚙️ Settings, then Danger Zone at the bottom of the screen) and create a fresh fork.

Where can I get help?

🙋 Students ask questions in the form of issues in the main repository, so that everyone can benefit. Make sure to include enough information such as screenshots and so on. Also, make sure you check the error logs of the automated tests as shown in the screenshots below.

Further Reading

• The Hitchhiker's Guide to Python
• One of your fellow students suggested this website