pcouy

joined 2 years ago
sorted by: new top controversial old
"New theory proposes time has three dimensions, with space as a secondary effect" -- now you get three detentions for being late to high school science class in c/[email protected]
[–] [email protected] 1 points 6 days ago

How about putting it at the top of the article, or using the dedicated "editor's note" button they put right below the title ?

"New theory proposes time has three dimensions, with space as a secondary effect" -- now you get three detentions for being late to high school science class in c/[email protected]
[–] [email protected] 1 points 6 days ago (2 children)

Found it thanks

Why did they burry what's arguably the most important piece of information at the very end though ?

"New theory proposes time has three dimensions, with space as a secondary effect" -- now you get three detentions for being late to high school science class in c/[email protected]
[–] [email protected] 1 points 6 days ago* (last edited 6 days ago) (4 children)

I'm not seeing this editor's note (at least on mobile). Where is it ?

Edit 2: never mind, found it

"New theory proposes time has three dimensions, with space as a secondary effect" -- now you get three detentions for being late to high school science class in c/[email protected]
[–] [email protected] 7 points 6 days ago (7 children)

How reliable is this website ? I see clickbaity headlines from it all the time around here and the Wikipedia page is mostly empty

Honeypot or just vv irresponsible database? in c/[email protected]
[–] [email protected] 1 points 1 week ago (1 children)

I'm not seeing a picture, and even then, we would not be able to give any insight from a picture alone

Honeypot or just vv irresponsible database? in c/[email protected]
[–] [email protected] 0 points 1 week ago (3 children)

First, your post is probably missing a link so we don't have any context on what you're asking (even if we can guess some stuff from the post text)

Second, you mention a website being sketchy/a honeypot without providing any technical reason to believe so

Third, this has nothing to do with computer security (maybe more of a privacy issue), and it does not look like a news piece, so this is definitely the wrong community

1
Livestream of my RL agent training on Super Mario Bros (twitch.tv)
 

publication croisée depuis : https://lemmy.pierre-couy.fr/post/1059609

Hey everyone!

I've been working on my own toy reinforcement learning (RL) framework for a while now and have nearly implemented a full Rainbow agent—though I'm still missing the distributional component due to some design choices that make integration tricky. Along the way, I’ve used this framework to experiment with various concepts, mainly reward normalization strategies and exploration policies.

I started by training the agent on simpler games like Snake, but things got really interesting when I moved on to Super Mario Bros. Watching the agent learn and improve has been incredibly fun, so I figured—why not share the experience? That’s why I’m streaming the learning process live!

Right now, the stream is fairly simple, but I plan to enhance it with overlays showing key details about the training run—such as hyperparameters, training steps/episodes, performance graphs, and maybe even a way to visualize the agent’s actions in real-time.

If you have any ideas on how to make the stream more engaging, or if you're curious about the implementation, feel free to ask!

Code Smells Catalog in c/[email protected]
[–] [email protected] 0 points 5 months ago (1 children)

Well it's in the name, they are code smells, not hard rules.

Regarding the specific example you cited, I think that with practice it becomes gradually more natural to write reusable functions and methods on the first iteration, removing the need for later DRY-related refactorings.

PS : I love how your quote for the Rule of Three is getting syntax highlighted xD (You can use markdown quotes by starting quoted lines with > )

0
Code Smells Catalog (luzkan.github.io)
1
Neural Network prototype on Shadertoy - Interactive handwritten digits classifier (www.shadertoy.com)
 

I've kept playing with shader programming and managed to export a trained neural network's weights as GLSL variable definitions. The code is ugly as hell as I've done a lot of quick experiments with it, and I went all-in with macros where functions would probably be better suited. I hope you still find it interesting.

Excluding neural network weights, the whole thing is ~300 lines of code and can run a few variations of a simple convolutional network.

1
Mitosis in the Gray-Scott model : an introduction to writing shader-based chemical simulations (pierre-couy.dev)
 

cross-posted from: https://lemmy.pierre-couy.fr/post/678825

Hi ! I've been working on this article for the past few days. It would mean a lot to me if you could provide some feedback.

It is about implementing a physico-chemical simulation as my first attempt to write a shader. The code is surprisingly simple and short (less than 100 lines). The "Prerequisite" and "Update rules" sections, however, may need some adjustments to make them clearer.

Thanks for reading

1
Mitosis in the Gray-Scott model : an introduction to writing shader-based chemical simulations (pierre-couy.dev)
 

cross-posted from: https://lemmy.pierre-couy.fr/post/678825

Hi ! I've been working on this article for the past few days. It would mean a lot to me if you could provide some feedback.

It is about implementing a physico-chemical simulation as my first attempt to write a shader. The code is surprisingly simple and short (less than 100 lines). The "Prerequisite" and "Update rules" sections, however, may need some adjustments to make them clearer.

Thanks for reading

0
‘Scanners are complicated’: why Gen Z faces workplace ‘tech shame’ (www.theguardian.com)