A Keras-trained RNN to emulate the works of a famous poet, powered by BrainCore
Swift Ruby
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
InfiniteMonkeys.xcodeproj
InfiniteMonkeys.xcworkspace
InfiniteMonkeys
InfiniteMonkeysApp
.gitignore
LICENSE
NetworkBuilder.swift
Podfile
Poet.swift
README.md
lstm_text_generation_weights.h5

README.md

InfiniteMonkeys

A Keras-trained RNN to emulate the works of a famous poet, powered by BrainCore. Contains targets for Mac OS and iOS, with a very basic UI.

What?

Go here first: https://karpathy.github.io/2015/05/21/rnn-effectiveness/.

And then here: https://github.com/aleph7/BrainCore.

Who?

The network has been trained for just a few hours on the works of Walt Whitman, obtained from Project Gutenberg.

You could improve the output by training for longer on a more powerful machine. The script required for training is from the Keras project, and you can find it here.

Sample output

and i will make the stands of the heart, and the sailors and the stars is string,
where the streets and rife of the earth is sorrows and the world, sure the streets them and life,
the steamy hand, and the sea of all,
the shores in the houses and the star, or the head, and are the stars is so spreading