1,615 karma · joined May 8, 2018
Reach me at: pgyurov93 at gmail dot com
It makes me wonder how much palaeolithic art has been lost because it wasn't done inside a cave.
I've been meaning to post about it.
https://github.com/petargyurov/virtual-bookshelf
And you can take a look at my blog repo to see how I've used it with Jekyll:
Well, 4 years later I decided to push a small update [1]. Ladies and gentlemen, we now have slanted books. The future is now.
[0] https://news.ycombinator.com/item?id=31293727
[1] https://github.com/petargyurov/virtual-bookshelf
PS: I need to read more.
Out of curiosity, how did you go about replacing the CUDA specific ops? Any resources you relied on or just experience? Would love to learn more.
To be honest I built it just for me and then decided it might be useful for others.
It's all local, no server, no database, etc. Mobile and desktop friendly.
It's all local, no server, no database, etc. Mobile and desktop friendly.
I did build it with the help of AI, but no AI inside the actual thing.
Yeah, it's more of a flashcard-crossword rather than a real crossword. I will have a think about how I can make the clues more interesting. I wanted to make this simple to use (and free) so adding an LLM in there might complicate things.
Thanks for trying it!
This is a tough problem to solve and for a lot of operators ultimately comes down to cost.
We're based in the UK but reach out if you wanna chat!
Also, I am not sure how heavy LIDAR units are, but remember that the heavier the payload the more the flight time is reduced. Some drones can only have a single payload, so if you also want to capture (high-res) video/imgs you need to fly again.
It all depends on the use-case.
I also rate the author's slogan: "make things, not too much, mostly crap". A reference to Michael Pollan's "eat food, not too much, mostly veg".
I remember one time when I was in the back and my dad took a rather sharp turn at a major junction and the rear door swung open. I calmly alerted him to the situation and with one hand on the steering wheel and one hand reaching behind, somehow got the door closed whilst I sat semi-afraid for my life.
I remember the horn also broke so he rewired it to a custom red button that he mounted on the driver's door handle.
No terms and conditions.
No thanks.
And that's on top of the already questionable product.
Regarding your README, I would add a text version of the example in the demo.
Whatever happened to that? Vapourware?
We'll get more incremental updates and nice features:
* more context size
* less hallucinations
* more prompt control (or the illusion of)
But we won't get AGI this way.
From the very beginning LLMs were shown to be incapable of synthesising new ideas. They don't sit there and think; they can only connect dots within a paradigm that we give them. You may give me examples of AI discovering new medicines and math proofs as a counter-argument but I see that as re-enforcing the above.
Paired with data and computional scaling issues, I just don't see it happening. They will remain a useful tool, but won't become AGI.
And whether they stay affordable is a question of time; all the big players are burning mountains of cash just to edge out the competition in terms of adoption.
Is there a level of adoption that can justify the current costs to run these things?
Seriously, who's idea was this? It can't be a money saving feature; surely it costs more to upscale all these videos than to just host the HD version.
And even if you argue it can be used only on low res videos to provide a "better experience", the resulting distortion of reality should be very concerning.
I wonder if 3D clustering would allow for more space.
But hey, another way to shoehorn AI into being "useful".
I'm sure someone will chime in with "skill issue", and that's true to some extent, but you gotta admit it's an over-engineered "solution".
Otherwise, Blender. Been using it at work for a left field innovation project and I'm kind of hooked.
Unlimited resource drain caused by conscious blobs of particles (a.k.a humans) invoking recursive thought patterns over their own existence.
Workaround: ponder function can return a feeling of existential dread accompanied by increased weights for headache "sensation", causing blob to terminate process.
Estimate: M
Label: Backend
At least say why you think so and contribute to the conversation a bit.
[0] https://news.ycombinator.com/newsguidelines.html#comments
Would you recommend Tamed?