tomtom1337

Great response! I'm just a layman here (former material scientist) but it's fun to think about this stuff!

You mean the argon gas used as medium specifically? I assume the purge gas is something else, cheaper?

Sorry, I guess. There were no comments here after a couple of hours and I literally felt bad, so I commented without being very exact in my writing. Regardless, great job, and thank you.

This is extremely cool reading! I'm impressed that they actually fined Elkjøp (as they should!) but very surprised that they didn't keep you informed!

Thank you for sharing!

This is literally what talaas has done with chatjimmy.ai.

Try it, it's llama 3.1 8B at 16000 tokens per second.

chatjimmy.ai https://taalas.com/the-path-to-ubiquitous-ai/

It is SO NICE to see people working on making fast, nice-to-use tools. It's a lovely experience to use diffshub. Thank you for creating it, and than you for the great write-up! (I made it "that far" )

Have you split your 100k loc codebases into smaller crates? If you take a look at eg gitoxide's repo, they've split it in many smaller crates. I think that might help with keeping the scope for the ai small and maybe help with keeping contracts tight and well-defined.

You could probably use just an X-scanner, and instead of a CCD line sensor, use a regular 2D image sensor if you used a "1 pixel wide" slit aperture to crop the image perpendicularly to the direction that the prism disperses the light. So instead of a single pixel being dispersed, you disperse a line.

You would reduce the time required by the root of the number of pixels you want (assuming a square image).

(This is what we do in momentum-resolved electron energy loss spectroscopy. In that situation we have electromagnetic lenses that focus the electrons that have been dispersed, so we don't have as bad a chromatic aberration problem as the other response mentions).

I would love to see e.g. a butterfly image with a slider that I could drag to choose the wavelength shown!!

This is called chromatic aberration, for those who are intrigued.

Given that regular phone cameras have sensors that detect RGB, I wonder if one could notice improved image sharpness if one had three camera lenses (and used single-color sensors) next to one another laterally, with a color filter for R, G and B for each one respectively. So that the camera could focus perfectly for each wavelength.

This is very interesting, but like sibling comments, I'm very curious as to how you run this in practice. Do you just tell Claude/Copilot to do what you describe?

And do you have any prompts to share?

I used it to map out storage locations and refill stations at our online grocery picking stations, then export it to read in using geopandas in order to calculate the shortest distances between all locations!

I get it but I think you're essentially saying "I learnt it this way and it's hard to change once I learnt it that way". My argument is that "when teaching others, it's better to teach them to use switch and restore rather than checkout".

I'd recommend using git switch instead of checkout, since the checkout command is so overloaded. And restore instead of checkout for restoring changes.

Definitely do not parse each row into eg pydantic models. You lose the entire performance benefit of pandas / polars by doing this.

If you need it, use a dataframe validation library to ensure that values are within certain ranges.

There are not yet good, fast implementations of proper types in Python dataframes (or databases for that matter) that I am aware of.

Any suggestions for «orchestrating» this type of experiment?

And how does one compare the results in a way that is easy to parse? 7 models producing 1 PR each is one way, but it doesn’t feel very easy to compare such.

You need something "more" on the website before you ask people to create an account. "Team workspace that stays fast" isn’t clear enough for me, at least. What is a workspace? What does the interface look like? Is it in the browser? Is it an app?

People will go "what is this?", "huh, I’m not gonna make a user for this, can’t tell what it is". Those are my 2 cents.

Haha, I started reading this, got interrupted, came back and got confused by the graph. Then came to the comments, saw your comment, reloaded the post and voila!

Thank you for a lovely post!

They mean «this kind of demand/leadership» no longer exists. Not the particular feature of the wake up time.

Oooh, thank you for sharing! New product lineup looks interesting, but I echo other concerns here about it thread maybe eventually requiring internet.

Are the ikea zigbee devices going to stop being sold? Massive shame if so, they are extremely reliable and easy to use.

Super quick feedback - opening that link on my phone shows me two options next to each other, seemingly with the same name / description (followed by …) and same pricetag. I had to turn my phone sideways to see that there is a windows and a Mac version.

I think you can afford the extra characters to show the whole page in portrait mode. (iPhone 16 pro Safari)

https://imgur.com/a/aTxO3sp

Could you provide a bit more context here? I’m looking at the math example (https://github.com/ohmjs/ohm/blob/main/examples/math/index.h...) and would like to learn a bit more.

If you’re young, new to data science and hoping to get a job in it after some time, then I absolutely recommend learning Python instead of Julia.

But if you are just interested in learning a new language and trying it in data science OR are not currently looking to enter the data science job market, then by all means: Julia is great and in many ways superior to Python for data science.

It’s just that «everyone» is doing data science in Python, and if you’re new to DS, then you should also know Python (but by all means learn Julia too!).

Look, this article is absolutely excellent, and answers your questions. Please read the article before commenting this sort of thing.

As someone who has had to use geopandas a lot, having something which is up to an order of magnitude faster is a real dream come true.

Ah, of course. Slightly ambiguous English tricked me there. Thank you Ritchie!

Out of curiosity, what makes a rust library easier to use? Could you expand on that?

Why would it be difficult to monitor the slowness? Wouldn’t a million function calls to the from_F_to_K function be very noticeable when profiling?

On your case about swapping between image representations: let’s say you’re doing a FFT to transform between real and reciprocal representations of an image - you probably have to do that transformation in order to do the the work you need doing on reciprocal space. There’s no getting around it. Or am I misunderstanding?

Please don’t take my response as criticism, I’m genuinely interested here, and enjoying the discussion.

I interpret your question as «given that I am doing many conversions between temperature, because that makes it easier to write correct code, then I worry that my code will be slow because I am doing many conversions».

My response is: these conversions are unlikely to be the slow step in your code, don’t worry about it.

I do agree though, that it would be nice if the compiler could simplify the math to remove the conversions between units. I don’t know of any languages that can do that.

Relevant xkcd: https://xkcd.com/2205/

You have a typo: In your last sentence you effectively wrote «from int to float» twice in contradicting ways. «To float from int than (…) from int to float».