tomtom1337

6 ms·

31.
by tomtom1337 1y ago
Ah, of course. Slightly ambiguous English tricked me there. Thank you Ritchie!
32.
by tomtom1337 1y ago
Out of curiosity, what makes a rust library easier to use? Could you expand on that?
33.
by tomtom1337 1y ago
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 t
34.
by tomtom1337 1y ago
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
35.
by tomtom1337 1y ago
Relevant xkcd: https://xkcd.com/2205/
36.
by tomtom1337 1y ago
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».
37.
by tomtom1337 1y ago
Could you expand on this? It sounds a bit preposterous to save a text, as json, inside an image - and then expect it to be immediately usable… as an image?
38.
by tomtom1337 1y ago
I suggest you wrap the misconceptions in quotation marks to make it clearer what is the misconception and what is not. Took me two passes to realize what was what.
39.
by tomtom1337 1y ago
I echo the other response here. You absolutely should switch to using notebooks in VSCode with their static typenchecker. Language Servers do exactly what you are wanting, with both notebook integration and «live coding».
40.
by tomtom1337 1y ago
Turn it off?
41.
by tomtom1337 2y ago
This resonates strongly! In Python, I often now find myself declaring Pydantic data structures, and then adding classmethods and regular methods on them to facilitate converting between them. It makes for great APIs (dot-chaining from one t
42.
by tomtom1337 2y ago
Which GPU Are you running on? I just got a 4080 SUPER and I’m thinking of trying Linux with it.
43.
by tomtom1337 2y ago
I use mise with uv for automatic activation of venvs when I cd into a directory containing one (alongside a mise.toml). Do you tackle this in some other manner?
44.
by tomtom1337 2y ago
Ive used this several times, I paid for pro. Really like it! Check out the hand recognition, it’s super cool (albeit, not terribly useful). There’s a very active and helpful community on discord too.
45.
by tomtom1337 2y ago
I didn’t understand this. What is it about the gamma factor that will cause the cosmic microwave background to evaporate you?
46.
by tomtom1337 2y ago
I would have liked to see the performance of the ash c version, seems like a surprising omission. I’m also confused about the perf2 performance. For the threads example it starts around 70_000 reqs/sec, while the processes example runs
47.
by tomtom1337 2y ago
Same on iPhone, Safari.
48.
by tomtom1337 2y ago
If you’re looking to analyse your hyperspectral images (spectrum-images, image-images or n-dimensional- n-dimensional datasets), I can highly recommend hyperspy [1]. One of the brilliant ideas hyperspy incorporates is that we consider datas
49.
by tomtom1337 2y ago
I enjoyed reading your solutions! I’m surprised how terse they are!
50.
by tomtom1337 2y ago
Rather than the snarky response, I’ll answer: closer is likely to be impossibly more expensive for the requirements OP needs (e.g. living area, number of bedrooms or similar).
51.
by tomtom1337 2y ago
Could you explain more what you mean? Other than certain holidays perhaps causing a higher chance of alone time for future mom and dad, I don’t see what you mean, and I don’t understand your last paragraph.
52.
by tomtom1337 2y ago
That’s an excellent example of Goodhart’s law! "When a measure becomes a target, it ceases to be a good measure".
53.
by tomtom1337 2y ago
Thank you for sharing! I am convinced! Completely agree that this is a problem!
54.
by tomtom1337 2y ago
I guess the unspoken assumption Babbage makes here is «if I put only the wrong figures into the machine». Then it is completely unreasonable to expect correct output. In ML context an LLM has been trained on much data, some «wrong» and so
55.
by tomtom1337 2y ago
I updated the benchmark [1] that I wrote to someone who responded to you to include a even/odd filter and take the minimum time of 100 runs. Python 3.13 is much faster for small lists, and a tiny bit faster for larger lists. The times
56.
by tomtom1337 2y ago
Great minds think alike. Here's one I just cooked up. Python 3.13 is faster, at least for the simplest possible list comprehension with no filters. N is the length of the list comprehension. Repeated 100 times for each test and showing
57.
by tomtom1337 2y ago
The animated ripple across the «what’s new» button is infuriating. It keeps drawing my attention from reading what this is.
58.
by tomtom1337 2y ago
Well, one single long exposure would literally generate a regular sized image. Though I do guess they’ll be doing multiple exposures if only to avoid saturating their dynamic range. Does anyone know what a typical number of acquired frames
59.
by tomtom1337 2y ago
The most common solution is to frequently retrain on the latest data. A forecasting model might retrain every week, including the last weeks data, and might even drop older data, for instance training data older than a year. It's best
60.
by tomtom1337 2y ago
I haven't read the text, but data drift refers to how, after deploying a machine learning model, the input data changes over time to something that wasn't tested on. For instance, let's say you create a gradient boosting fore

More ›