Computer Setup and Software Installs
This course run is entirely using Jupyter Notebook. It is an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and narrative text. We will therefore only be using the terminal window on your computer for the initial setup and software installation.
Terminal Setup
Linux: Look for Terminal in your applications and click on it to open it.
macOS: Bash is the default shell in all versions of macOS, you do not need to install anything. Open the Terminal application from /Applications/Utilities or using Spotlight search.
Windows: Open the Start Menu, type command or cmd in the search field, and then press Enter or click on the Command Prompt shortcut.
Software Installs for Day 1
For this course, we ask you to install Miniconda and as well as the a series of software packages with it (jupyter notebook, nltk, numpy, matplotlib, wordcloud, pandas and networkx). We also ask you to install the Stanford NER tagger (which is needed for day 2 of the course). See instructions below. Please take time to install these different things prior to the course.
1. Install Miniconda
Got to https://docs.conda.io/en/latest/miniconda.html and download the correct version. We will be working with Python version 3.7, so you need to select the appropriate version of Miniconda for your operating system with Python 3.7.
For all of the software packages to be installed next (using conda) you will need to proceed through the terminal installation by pressing “y” when prompted on Linux/Mac or install them via the Anaconda prompt on Windows.
2. Install Jupyter Notebook
Windows: install jupyter notebook via the Anaconda prompt
macOS/Linux: go to a terminal window and run the following command:
conda install -c anaconda jupyter
3. Install NLTK
Windows: install NLTK via the Anaconda prompt
macOS/Linux: run the following command in your terminal:
conda install -c anaconda nltk
4. Install numpy
Windows: install numpy via the Anaconda prompt
macOS/Linux: run the following command in your terminal:
conda install numpy
5. Install matplotlib
Windows: install matplotlib via the Anaconda prompt
macOS/Linux: run the following command in your terminal
conda install -c conda-forge matplotlib
6. Install wordcloud
Windows: install wordcloud via the Anaconda prompt
macOS/Linux: run the following command in your terminal:
conda install -c conda-forge wordcloud
Software Installs for Day 2
7. Install Stanford NER tagger
Go to the following URL and if your computer doesn’t automatically download it, then download it yourself.
https://nlp.stanford.edu/software/stanford-ner-2018-10-16.zip
You can put the tagger anywhere you like. For this workshop we recommend you put it into your Download folder. The directory path to the Stanford tagger is needed to be able to run some of the exercises on day 2.
8. Install pandas
Windows: install pandas via the Anaconda prompt
macOS/Linux: run the following command in your terminal:
conda install pandas
9. Install networkx
Windows: install networkx via the Anaconda prompt
macOS/Linux: run the following command in your terminal:
conda install networkx