Yeah I think a local Git server would be good, will try our forgejo since people seem to like it— I’ve been using git for a lot of projects but not so much for large files and HW stuff since when using GitHub there are size limitations. Does seem like it would be freeing to be able to delete whatever I want from my workstation without worrying about losing stuff

Def going to check this out— I think it will solve my biggest problem with GitHub which is just the number of large files I have to deal with— overthought comments are the best comments- thanks!

Yeah, I really should start using Git for everything, but I’ve been working with a lot of large datasets recently (mostly EEG data). A big part of improving accuracy comes from cleaning the data, which is huge and takes a while to process. I could set up a local Git server to keep track of everything or just save the base data files and regenerate as needed, but on my current setup, that process can take anywhere from 2-6 hours depending on the task. So for now, I’ve just been managing everything locally to save time.

IRC: it’s open source, it’s free, its retro

But Linux is open source? So if hypothetically so distro adopted spying al la windows couldn’t people just change distros? tbh I also think the question is slightly confusing as I don’t understand why OP thinks Mac OS is not standardized but I digress.

I like your optimism best to look on the bright side and all— curious what do you mean by fabless? Do they not require as complex facilities because they’re a larger process or something? Or for some other reason?

deleted by creator

For fucks sake I hope your joking x.x

Thanks for the tip! https://github.com/jcuberdruid/DeepBlue

I’ll add more to the repo but here it is for now :)

Not just yet but it’s in the works!

Thanks! I made a submarine game called DeepBlue where you explore the ocean floor generated by ocean depth map data:

Millions of hard working Dutch people depend on the tulip system for their retirement. The crash was not due to the management of the tulip market, rather it’s an unavoidable pattern of boom and bust. Given this, it’s only right that the Dutch government to bail out the tulip funds to avoid a breakdown of the liliaceae system.

While the iris study is interesting, looking at their dataset the pictures seem to include the area around the eye a little bit, including eye lashes, so after a cursory glance it seems odd that they even titled it as iris. However I didn’t read the full thing so it cold be that they cropped it somewhere. Although they are using large convolutions so a lot of detail is lost.

I do really like the error system in rust for its descriptions. I guess the difficulty for me, which maybe will go away after writing more rust, is that my intuition for what is efficient and what isn’t totally breaks down.

I find myself passing copies of values around and things like that, it might be that the compiler just takes care of that, or that I just don’t know how to do it well but that’s often the point of friction for me.

Totally agree on the refactor though, most of the time it doesn’t even take that much time since you know the skeleton of what you want at that point!

Maybe it’s just because I haven’t had to deal with the scenario yet but does compile time really matter? I mean for small programs it seems it’s almost instant on modern machines and for large programs I would assume, if it exists, that you would be using the equivalent of make so you would only be recompiling the small changes made.

deleted by creator

Llama2 is pretty good but there are a ton of different models which have different pros and cons, you can see some of them here: https://ollama.com/library . However I would say that as a whole models are generally slightly less polished compared to chat gpt.

To put it another way: when things are good they’re just as good, but when things are bad the AI will start going off the rails, for instance holding both sides on the conversation, refusing to answer, just saying goodbye, etc. More “wild westy” but you can also save the chats and go back to them so there are ways to mitigate, and things are only getting better.

Also Swift!

While I agree wholeheartedly with the idea that we need to emphasize quality over quantity, so long as software pays well there will be people who don’t care. In my university I’ve met a fair few people that complain about having to learn about compilers, assembly, and whatnot because “I’ll never need to know that in my actual job”. While to some extent in the United States you can blame the fact that classes just cost a ton, I think it’s a sad reality that, barring some key change in the way our whole education and economic systems work, there will be unimaginative apathetic people that will ruin things for the rest. Plus people are fallible or something I dunno. But yeah void pointers are my jam because I don’t have to wait precious clock cycles making new ones jk.

Just run ollama locally and download uncensored versions— runs on my m1 MacBook no problem and is at the very least comparable to chatgpt3. Unsure for images though, but there should be some open source options. Data is king here, so the more you use a platform the better its AI gets (generally) so don’t give the corporations the business.