Markov chain #

this is a simple markov chain implementation inspired by a project @nathansolomon1678 made

Generate chain #

The text used to make the chain should be in ./all_stuff.txt, or change the string in markov.py

from markov import MakeMarkov

mark = MakeMarkov()

# Only needed on first run
mark.setup_db()

# Pull all the words from the text file
mark.read_file()

# Make the sql database of unique words, only needed on first run
mark.find_unique()	# This will output to a database called `markov.db`

# Finally build the markov chain!
mark.train()	# This will output to a file called `markov.json`

Generate text (in code) #

from markov import generate

# Some text will be generated with 100 words
text = generate()

# Some text will start with the word "The"
text = generate(start="The")

# Some text with the length of 1000 words
text = generate(length=1000)

# Some text with the length of 1000 words and starting with "The"
text = generate(length=1000, start="The")

Generate text (with api) #

uvicorn fast:app --reload

Some text will be generated with 100 words #

http://127.0.0.1:8000

Some text will be generated with 1000 words #

http://127.0.0.1:8000?length=1000

Some text will start with the word “The” #

http://127.0.0.1:8000?start=The

Some text will start with the word “The” and with the length of 1000 #

http://127.0.0.1:8000?start=The&length=1000