This is an attempt to explore the possibility to use Processing as a tool to incroperate more human inputs

These scribbles were made using a special “bursh” was constrained to only go diagonally following the directioin of the cursor

I quite like that sheep dog!

Can machines write poetry?

In this attempt, a GTP-2 neuron network was trained with 2618 lines of Shakespeare’s Sonnets.

Because the input data is relatively small, the output was slightly overfited. The 4 output samples here were cherrypicked.