ipsum2

52 ms·

1. 781.▲
   by ipsum2 4y ago

   It's only hypocrisy if you think that HN is made up of a single person commenting under multiple accounts and not a diverse group of people with varying opinions.
2. 782.▲
   by ipsum2 4y ago

   This is amazing. Out of curiosity, why not try deep learning OCR software instead of Tesseract? PaddleOCR is popular.
3. 783.▲
   by ipsum2 4y ago

   Thanks for the correction, I meant EFA but misremembered the name.
4. 784.▲
   by ipsum2 4y ago

   With AWS's elastic network adapter (ENA) providing fast interconnect, there's practically not much difference, assuming you can purchase the total amount of compute.
5. 785.▲
   by ipsum2 4y ago

   Actually, Intel beats M1/M2 in perf/watt benchmarks: https://www.cpubenchmark.net/power_performance.html
6. 786.▲
   by ipsum2 4y ago

   Can you expand on the inconsistencies between what the model predicts and reality? I'd love to learn more and see some examples.
7. 787.▲
   by ipsum2 4y ago

   How is codejumper $1k for something so simple? It doesn't seem very inclusive, and it could be easily DIYable.
8. 788.▲
   by ipsum2 4y ago

   @stevenpetryk do you see Mafs becoming a library supporting similar features to Manim?
9. 789.▲
   by ipsum2 4y ago

   Stearic acids are used everywhere, including detergents, shampoos, soaps, etc. TIL they're made from tallow.
10. 790.▲
    by ipsum2 4y ago

    This is a book about US cuisine?
11. 791.▲
    by ipsum2 4y ago

    It's pretty clear from the article that the author defines supplier as being in a business relationship, using "supply chain" as the context.
12. 792.▲
    by ipsum2 4y ago

    I've updated my comment to clarify the question
13. 793.▲
    by ipsum2 4y ago

    Forgive me if I'm misunderstanding how it works, is the model is predicting the original emoji from an augmented emoji, and doing that for every 36x36 patch in the webcam image?
14. 794.▲
    by ipsum2 4y ago

    Possibly, though I don't know of any companies that are training ML models on laptop hardware.
15. 795.▲
    by ipsum2 4y ago

    You can buy the exact same laptop, but $1300 cheaper if you can install Linux yourself: https://www.razer.com/gaming-laptops/Razer-Blade-15/RZ09-042... Or get a more powerful CPU for $100 cheaper than the Razer wi
16. 796.▲
    by ipsum2 4y ago

    It's not going to take off, since it doesn't have full (or even most) API compatibility with Python. Numba seems strictly better because it can interop with Python.
17. 797.▲
    by ipsum2 4y ago

    Newbie question: how is this different than using an Arduino + power injection?
18. 798.▲
    by ipsum2 4y ago

    DAIN[0] is very popular for increasing framerate of videos, including animations. Examples: https://www.youtube.com/watch?v=IK-Q3EcTnTA https://www.youtube.com/watch?v=unRdyQYwL5k https://www.you
19. 799.▲
    by ipsum2 4y ago

    But chatGPT can't connect to the internet, it said so itself. Sorry, couldn't resist pointing out this garden path title.
20. 800.▲
    by ipsum2 4y ago

    Thanks for the correction.
21. 801.▲
    by ipsum2 4y ago

    Twitter is great, but its the users that provide the value, not Twitter itself. If all the interesting people are moving to Mastodon, guess I'll need to move as well.
22. 802.▲
    by ipsum2 4y ago

    Doesn't Accenture offshore a lot of its work to India and other third-world countries? So the Accenture engineer probably gets paid a lot less than you. Edit: According to Wikipedia, yes: > In 2015, the company had about 150,000 emp
23. 803.▲
    by ipsum2 4y ago

    It's just "enforcement", there's nothing malicious about it. Elon Musk is the CEO and policymaker of Twitter now, he's responsible for the actions his company takes.
24. 804.▲
    by ipsum2 4y ago

    It's news to me that that users are not allowed to mention other social networks' accounts on Twitter anymore. Seems short sighted, how many users is Twitter losing to Instagram/Discord/Mastodon?
25. 805.▲
    by ipsum2 4y ago

    I heard a 3 year old play the violin once, I'm very bearish on music.
26. 806.▲
    by ipsum2 4y ago

    That technically counts, but they're not really lending any expertise, just a few thousand to pay for the intern.
27. 807.▲
    by ipsum2 4y ago

    Isn't the connection that Maps runs ads?
28. 808.▲
    by ipsum2 4y ago

    Microsoft and Meta have contributed a ton of data to OSM, but I can't find any sources about Google supporting OSM.
29. 809.▲
    by ipsum2 4y ago

    You seem knowledgeable in this area. Do you know how image sensor support works? Can you attach any camera with CSI to the SBCs that you listed?
30. 810.▲
    by ipsum2 4y ago

    Looks like they have 4 different tokenizers. Besides gpt2, does anyone know which models correspond to which tokenizers? One of them is probably Codex. "gpt2": gpt2, "r50k_base": r50k_base, "p50k_base&