Skip to content

Latest commit

 

History

History
45 lines (30 loc) · 3.27 KB

README.md

File metadata and controls

45 lines (30 loc) · 3.27 KB
Raw

Topics – Easy Topic Modeling in Python

Topics is a gentle introduction to Topic Modeling. It provides a convenient, modular workflow that can be entirely controlled from within and which comes with a well documented Jupyter notebook, integrating three of the most popular LDA implementations: Gensim, MALLET, and lda. Users not yet familiar with working with Python scripts can test basic topic modeling in a Flask-based GUI demonstrator.

At the moment, there are three Jupyter notebooks:

Getting Started

Windows

  1. Download and install the latest version of WinPython.
  2. Download and install Git.
  3. Open the WinPython PowerShell Prompt.exe in your WinPython folder and type git clone https://github.com/DARIAH-DE/Topics.git to clone Topics into your WinPython folder.
  4. Type cd .\Topics in WinPython PowerShell to navigate to the Topics folder. 5a. Either: Type pip install . in Winpython PowerShell to install packages required by Topics 5b. Or: Type pip install -r requirements.txt in Winpython PowerShell to install Topics with additional development packages.
  5. Type jupyter notebook in WinPython PowerShell to open Jupyter, select one of the files with suffix .ipynb and follow the instructions.
  6. Note: For the development packages the Python module future is needed. Depending in your WinPython and your Windows version you might have to install future manually.
  7. Therefore, download the latest future-x.xx.x-py3-none-any.whl.
  8. Open the WinPython Control Panel.exe in your WinPython folder.
  9. Install the future-wheel via the WinPython Control Panel.exe.

macOS and Linux

  1. Download and install Git.
  2. Open the command-line interface, type git clone https://github.com/DARIAH-DE/Topics.git to clone Topics into your working directory.
  3. Note: The distribution packages libfreetype6-dev and libpng-dev and a compiler for C++, e.g. gcc have to be installed.
  4. Open the command-line interface, navigate to the folder Topics and type pip install . --user to install the required packages.
  5. Install Jupyter and run it by typing jupyter notebook in the command-line.
  6. Access the folder Topics through Jupyter in your browser, select one of the files with suffix .ipynb and follow the instructions.

Working with MALLET

  1. Download and unzip MALLET.
  2. Set the environment variable for MALLET.

For more detailed instructions, have a look at this.