Skip to content

Markov chain-based word generator based on Sam Twidale's implementation in Haxe.

License

Notifications You must be signed in to change notification settings

bicobus/pyMarkovNameGenerator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

https://img.shields.io/pypi/pyversions/markovname

pyMarkovNameGenerator

Markov chain-based name generator based on Sam Twidale implementation in Haxe. consider this work to be a python port with functionnally the same level of functionality. Feed it a list of training data, and it'll spew out a bunch of new words.

It comes with the training data compiled by Sam, and made available under Creative Commons BY-NC 4.0.

If you appreciate this work, consider buying me a coffee.

Install using:

pip install markovname

Requirements

This software has been developped and tested with python 3.9. Any earlier version might not work.

A minor dependency on wcwidth, make sure it is installed using pip, your package manager or whatever method you are familiar with. It is only necessary if you plan to use the main module. If you're only interested in the generator, then wcwidth can be ignored.

If you have poetry available, you can initialize a virtual environment with:

poetry install

Usage

The software reads sample data from json formatted datasets. See the data folder of this repository for examples.

usage: __main__.py [-h] [--list] [-n NUMBER] [-o ORDER] [-p PRIOR] [-g KEY] [-V] DATA

positional arguments:
  DATA                  Path containing the data sets.

optional arguments:
  -h, --help            show this help message and exit
  --list                List the various dataset available to the software.
                        (default: False)
  -n NUMBER, --number NUMBER
                        Amount of names to generate. (default: 1)
  -o ORDER, --order ORDER
                        Highest order of model to use. Will use Katz's back-
                        off model. It looks for the next letter based on the
                        last "n" letters. (default: 3)
  -p PRIOR, --prior PRIOR
                        The prior adds a constant probability that a random
                        letter is picked from the alphabet when generating a
                        new letter. Must be a number between 0 and 1.
                        (default: 0)
  -g KEY, --generate KEY
                        Which dataset to use. If not specified, use a random
                        available set instead. (default: None)
  -V, --version         show program's version number and exit

About

Markov chain-based word generator based on Sam Twidale's implementation in Haxe.

Topics

Resources

License

Stars

Watchers

Forks

Sponsor this project

Packages

No packages published

Languages