Comments on: Python Evolution – part 1

By: Evolving pseudo-creatures in computer simulation to run a short course » Lousy Canuck

Sun, 28 Apr 2013 04:39:33 +0000

[…] an image with random mutations, and some of the less-adapted would randomly be killed. And I showed even more lenient selection criteria work. I also noticed a sort of punctuated equilibrium, sometimes resulting in evolutionarily novel and […]

By: Python Picture Evolution part 2, finally | Lousy Canuck

Python Picture Evolution part 2, finally | Lousy Canuck — Fri, 09 Sep 2011 18:42:01 +0000

[…] I’ve finally gotten around to working on my picture evolution project. part 1 being here. You can feed it a sample image, tweak a bunch of parameters (at the top of the headers), then sit […]

By: Lousy Canuck » Python Picture Evolution part 2, finally

Lousy Canuck » Python Picture Evolution part 2, finally — Wed, 01 Apr 2009 23:06:03 +0000

By: jthibeault

jthibeault — Tue, 13 Jan 2009 01:56:02 +0000

In reply to scikidus.

Oops! I misremembered your app as being Java based. My apologies, I’ll fix the link.

I remember getting to your app from a link in a Youtube video where a program is described that does much the same thing, only with the color being a shade of grey from 00 to FF. I’m guessing you built on that idea, just like I’m trying to build on the EvoLisa idea — we can only see farther than our predecessors because we stand on their shoulders, and all that. Hat tip to you, sir.

Progress is stalled mostly due to work (again) getting in the way of the really important stuff, like blogging and coding random Python apps. By all means, grab the source and set ‘er up. Many eyes make all bugs shallow. Python is a very simple to understand and learn, and a very freeing overall, language — if you could code evolution in ActionScript, I’m sure learning Python will be a piece of cake.

By: scikidus

scikidus — Tue, 13 Jan 2009 01:24:11 +0000

Hi there! I’m glad to see that you enjoyed my little flash application. I was slightly unlucky releasing it around the same time as the EvoLisa project, but no worries. I wrote all of the code from scratch, and from your post it seems that you’re having trouble on my sid eof code. I don’t know Python, but I’m willing to help out in any way I can. Feel free to e-mail me if you hit a roadblock. 🙂

scikidus

By: jthibeault

jthibeault — Mon, 12 Jan 2009 17:18:26 +0000

In reply to André Roberge.

I’m using Pygame, drawing each polygon onto one surface then alpha-blitting them onto the main surface. It’s definitely a slow approach, as I’m getting maybe 40 iterations a second on my laptop. I haven’t tried yet on my much-beefier desktop.

Speaking of which, I have the source code now — I’ll edit this post shortly to include it.

By: André Roberge

André Roberge — Mon, 12 Jan 2009 17:07:03 +0000

What do you use to draw? I’ve started on a Python version myself using aggdraw and PIL and get about 70 mutations per second (for a 200×200 pixel image) – I’d be curious to hear about the speed obtained using different approaches as they have an impact on exploring various algorithms or variation on a theme (like what you are apparently doing).