Open rupeshs opened 7 years ago
Nice work! We were thinking about writing something like this. Good to see someone has done it.
Thank you
That's fantastic! I've added a link to your demo in README.
That is cool...thanks
On Jan 4, 2017 11:30 AM, "Xinyu Zhou" notifications@github.com wrote:
That's fantastic! I've added a link to your demo in README.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/zxytim/neural-painter/issues/2#issuecomment-270300572, or mute the thread https://github.com/notifications/unsubscribe-auth/ADGuug7GcZYbX7CiZVi8ADc4AcMeKynYks5rOzWKgaJpZM4LaTU7 .
BTW, there's a --recurrent
option that enable parameter sharing among layers in 'neural_painter.py', turning the MLP into a RNN.
However, the results are not much different like a MLP at rough glance.
https://rupeshs.github.io/npainter source https://github.com/rupeshs/npainter