TextFlow

Summary

A general purpose web application for generating text from various sources (books, academic papers, tweets, facebook posts, and reddit posts). The text will match the semantic style of the author(s) selected by the user using markov chain text generation techniques.

Demo

https://www.youtube.com/watch?v=tTQXbw1gR-o

Important files:

app.py

This is the file that runs the server

markov_final.py

This is the file that contains the python code for the Markov text generation algorithm. The function takes the following three arguments: source: The input text source that the generated text will be generated from length: The number of words to generate lookback: The number of words used as a key to "predict" the following word

Important commands:

source ../bin/activate

This starts the virtual environment.

gunicorn --bind 0.0.0.0:8080 app:app &

Runs the gunicorn server Needs to be run whenever app.py changes

sudo nano /etc/nginx/nginx.conf

Edits the nginx config file

sudo service nginx start

sudo service nginx stop

Needs to be run whenever the conf file changes

ps ax|grep gunicorn

Lists the gunicorn processes

kill <process_id>

Shuts down the gunicorn server

Languages & tools

Python flask, HTML, CSS, Bootstrap, JavaScript, jQuery, SQLite

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
scrapers		scrapers
static		static
templates		templates
Capture.JPG		Capture.JPG
README.md		README.md
app.py		app.py
app.pyc		app.pyc
db.py		db.py
db.pyc		db.pyc
documents.db-journal		documents.db-journal
generator.py		generator.py
generator.pyc		generator.pyc
get_probability.py		get_probability.py
markov.py		markov.py
markov.pyc		markov.pyc
markov_final.py		markov_final.py
old_markov.py		old_markov.py
populate.py		populate.py

lukeymyuan/TextFlow

Folders and files

Latest commit

History

Repository files navigation