Coder Social home page Coder Social logo

ivanatrumpalot's Introduction

Status Neural net version

Make America Hate Again

@IvanaTrumpalot

A deep learning bot trained on the crap coming out of Donald Trump's mouth.

Wait, what now ?

Ivana Trumpalot is an Ivana Trumpabot that uses a deep neural network to generate text that sounds like Donald Trump's speeches. It's currently not quite online, we're putting the finishing touches to it. We'll be continually training it over the near future to improve its Trumpiness.

Bot Functionality

Got another Twitter user in mind ? Let us know by opening up an issue.

Network

Ivana Trumpalot is a recurrent neural network using the Long Short-Term Memory paradigm. The code is written in Keras, and is fully available in this repository.

Data

We downloaded a handful of Trump's speeches, purged them for special characters we didn't want to keep, and removed any audience participation that was transcribed the the speeches. The corpus currently consists of about 250,000 characters, or 46,000 words. This corpus is also available in this repo.

Hardware

The live version of Ivana Trumpalot runs off a Raspberry Pi 3, where text is generated and tweets are sent to Twitter. The network was trained on a GTX 980 Ti, where an epoch takes about three minutes to run at the current architecture.

Usage

Training : To train the model, simply call code/train_lstm.py. Feel free to change the parameters and architecture of the network, as well as the dataset you use. The network will be saved to disk every five epochs.

Predictions : Using the predict() function in code/ivanatrumpalot.py, feed it some text and it will generate some output for you.

Twitter : If you want to set up a Twitter bot, we use this package. Don't forget to feed your API keys to code/twitter_bot.py.

The Name

Ivana Trumpalot is, of course, a reference to the antagonist in Austin Powers 2 : The Spy Who Shagged Me. She's groovy, baby. Ivana Trump is also Donald Trump's wife. This isn't a personal attack at her; you must understand, the name was simply too great to pass up. We have the best names. The best names.

The Authors

Mike and Quentin are as interested in deep learning and neural nets as Trump is interested in building big walls. They did their PhDs at the University of Warwick, and are both currently working in Vancouver, BC.

ivanatrumpalot's People

Contributors

qcaudron avatar sempwn avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.