pymarkoff 0.4.3

A simple Markov chain modeller and generator for word and sentence generation.

Originally a proof of concept, I’ve used this in enough projects that I’ve decided to publish it tomake it easier to import. The name is a play on words similar to Markup/Markdown. Hey, I’m funny sometimes,

Basic Use

Use the included from_sentences() and from_words() if your data already behaves nicely. These methods return a Markov object which is the workhorse of this whole thing. Your input sequences of words or sentences should be delimited by newlines. Use next_word() and next_sentence() to generate your desired forat of output.

Advanced Use

Instantiate with m = markoff.Markov(seeds) where seeds is an iterable of sub-iterables. That could be a list of lists of words or letters if you want to emulate the functionality of from_sentences() and from_words(). Each sub-iterable being a chain in the set of chains you want to model.

You can supply it with just one chain or many.

Then use m.generate(max_length=100) to produce a single chain limited to max_length automatically terminating at known ending state (usually a word ending with punctuation).

Examples

Don’t forget to begin with import pymarkoff ## Generating Sentences ##

Input

m = pymarkoff.from_sentences(
"""The quick brown fox jumped over the lazy dog.
Jack and Jill ran up the hill to fetch a pail of water.
Whenever the black fox jumped the squirrel gazed suspiciously."""
)

print([m.next_sentence() for i in range(10)])

Output

[
    'The quick brown fox jumped over the black fox jumped the lazy dog.',
    'The quick brown fox jumped the squirrel gazed suspiciously.',
    'Whenever the squirrel gazed suspiciously.',
    'Jack and Jill ran up the lazy dog.',
    'Jack and Jill ran up the hill to fetch a pail of water.',
    'Jack and Jill ran up the black fox jumped the hill to fetch a pail of water.',
    'Whenever the lazy dog.',
    'The quick brown fox jumped over the lazy dog.',
    'Jack and Jill ran up the hill to fetch a pail of water.',
    'Jack and Jill ran up the squirrel gazed suspiciously.'
 ]

Generating Words

Input

seeds = """Ana
Bastion
D.Va
Genji
Hanzo
Junkrat
Lúcio
McCree
Mei
Mercy
Pharah
Reaper
Reinhardt
Roadhog
Soldier: 76
Symmetra
Torbjörn
Tracer
Widowmaker
Winston
Zarya
Zenyatta"""

brain = pymarkoff.from_words(mystr)
print([brain.next_word() for i in range(10)])

Output

['Zen', 'D.Vaperein', 'Za', 'To', 'Merya', 'Metrdo', 'So', 'Junjör', 'Ph', 'Mera']

More advanced use

The Markov.feed() method can be used to add more data into the model of a Markov object. This lets you add sentences or words one at a time, if you want.

Notes

This module is still under development and is mostly for me to play around with and learn Markov Chains. Cheers.