TuLeD-phylo - tupi-guarani

Code and data for classification of Tupi-Guarani languages

System requirements

The analysis pipeline has a number of software requirements, mainly a working Python installation, for preparing the data, and a Java environment for running beast2.

The minimum Python version is 3.7. You can verify if your version of Python is compatible by issuing the following command:

$ python --version

Please note that in some systems the executable might be called python3, python3.7 or something similar. If you don't have Python installed, you can follow these instructions.

The system also requires the pip package manager, which should be already installed in most systems. Once more, you can check with the pip --version command; if you do not have pip installed, find the appropriate way to install it in your system (which might be different from the official instructions).

You also need to have git installed and working. If you don't, consult the documentation for your system (Bitbucket provides a good tutorial).

It is highly recommended that you use a virtual environment. Different Python distributions, such as Anaconda, will offer different alternatives for virtual environments, but here we will follow the most common one.

(instructions for Java)

Set up the repository

Go to the directory where you want to store these data, clone this repository and move into it:

$ git clone https://github.com/LanguageStructure/phyloTG
$ cd phyloTG

Create a virtual environment and activate it:

$ python -m venv env
$ source env/bin/activate

Remember that, once you are done, you can leave the virtual environment with the deactivate command.

You should also make sure that pip and related libraries are up-to-date:

$ pip install --upgrade pip wheel

Install the necessary Python packages:

$ pip install -r requirements.txt

If you have never installed the CLDF catalogs before, issue the following command, which will set up a system-wide copy of the GitHub repositories for Glottolog, Concepticon, and CLTS. Note that this can take some minutes depending on your connection speed.

$ cldfbench catconfig -q

Install BEAST

(...)

Download released data

The following Python script will download the latest released data, storing it in raw/, and uncompress its contents to data/.

$ python download_data.py

Build the BEAST2 XML configuration

To build the BEAST2 XML configuration, using beastling, run:

$ python build_xml.py

Run the analysis

To run the analysis with BEAST, which takes a long time, run:

$ beast phylotupi-expert.xml