m3at

7 ms·

31.
Argentina to devalue peso by over 50% as part of emergency economic reforms(cnn.com)
2 points by m3at 3y ago|1 comments
32.
by m3at 3y ago
For others that were confused by the Gemini versions: the main one being discussed is Gemini Ultra (which is claimed to beat GPT-4). The one available through Bard is Gemini Pro . For the differences, looking at the technical report [1]
33.
by m3at 3y ago
I thought so too but no, Gemini _Ultra_ is supposed to beat GPT-4. The one available through Bard is Gemini Pro. The wording sure is confusing! Looking at the technical report [1], on selected benchmarks, rounded score in %: Dataset | Gemin
34.
by m3at 3y ago
I can recommend mindustry as well: https://mindustrygame.github.io/
35.
by m3at 3y ago
This is not an embedding model though. Yes you can always extract some embeddings from somewhere, but for most LLMs those won't perform well for retrieval (which makes sense as it's not what the models are optimizing for)
36.
by m3at 3y ago
To take an other example, would you call a game that has its code and all assets (ex. character sprites) freely available open source? Or would the process that was used to create the assets in the first place also be required to be conside
37.
Pulsar navigation: piloting aircraft with the aid of the stars(extremetech.com)
2 points by m3at 3y ago|0 comments
38.
by m3at 3y ago
I would only recommend pgvector if you're already primarily relying on postgres and the scale is limited (<1M documents). It won't handle the part that generates the embeddings though. You could use cloud vendors if you'
39.
by m3at 3y ago
Depends what you mean by taking all your articles. If it's scraping no, but if you provide text content and urls as key/values pairs yes. SageMaker and VertexAI are the AI services of AWS and GCP respectively, and they both offer
40.
Building a baseline JIT for Lua automatically(sillycross.github.io)
1 points by m3at 3y ago|0 comments
41.
by m3at 3y ago
I learned about this effect too recently in an ACX post, let me add a link for the curious: https://www.astralcodexten.com/p/your-book-review-secret-gov...
42.
by m3at 3y ago
I was surprised too! While I could find some excuses to exclude ARIMA, notably that in practice you need to input some important priors about your time series (periodicity, refinements for turning points, etc) for it to work decently, &quot
43.
Drawing with Code(kyndinfo.notion.site)
1 points by m3at 3y ago|0 comments
44.
by m3at 3y ago
Thanks for sharing, your mission is highly interesting! The DS lead role could be an ideal match, though if you allow me to be blunt to make the best of both of our time: is there flexibility in the package? If yes I'll reach out, or f
45.
by m3at 3y ago
Those prefix are great and I use them regularly! Though imo the killer feature of FF address bar is simply that it's tied to a proper search history . Unlike chrome (which I sadly have to use at work), that only keep 90 days of histor
46.
by m3at 3y ago
Good pitch and I appreciate that your landing page itself is in the spirit of what you do! > heavy R&D in AI / LLM / GPT Can you expand on your plans in the AI space? I'm full stack for everything ML related with 7 yea
47.
by m3at 3y ago
https://paulw.tokyo/ Few posts a year, mainly about machine learning
48.
by m3at 3y ago
I think it's hard to come up with a regex that'll cover everything, but that's the kind of things GPT is good for. I made put some scripts here, works well enough: https://github.com/m3at/hn_jobs_gpt_etl
49.
by m3at 3y ago
Very interested, will send you an email shortly
50.
by m3at 3y ago
You can find some demos linked on github [1], the simplest being: https://demo.mcaptcha.org/widget/?sitekey=pHy0AktWyOKuxZDzFf... [1] https://github.com/mCaptcha/mCaptcha#client-side-widget
51.
Make your own VPN with Fly.io, tailscale and GitHub(github.com)
332 points by m3at 3y ago|143 comments
52.
by m3at 3y ago
I just tried well word, unlike a sibling comment I think it's fun :) Though I have to agree that it's a bit unclear how to proceed at first. Maybe make a link to the tutorial more easily accessible, or even auto-suggested on first
53.
by m3at 4y ago
Nice work! I've made a similar demo as a web app [1] and obtained a few orders of magnitude faster performance, in the 50ms for 1.6M images on single laptop core. I've limited experience with coreml but there's a few tricks t
54.
by m3at 4y ago
> By default Brave will not block first Party ads which are actually part of the sites you visit https://support.brave.com/hc/en-us/articles/360022806212-How...
55.
by m3at 4y ago
The other way around
56.
by m3at 4y ago
Will do if I have feedback :)
57.
by m3at 4y ago
Now I realise that I _somehow_ totally missed your link to Page's work right in the README… Thanks for the detailed comment, I definitely will dive into your script soon! One early suggestion, you might want to try torch dynamo [1], an
58.
by m3at 4y ago
Nice experiment! Not to minimize your work, in fact some ideas might be complementary to yours, a couple years ago there were already some approaches reaching 30s on a single V100: https://myrtle.ai/learn/how-to-train-y
59.
by m3at 4y ago
Agreed. It's maybe like Linux versioning scheme, where 6.0 is just the one that comes after 5.19: > The major version number is incremented when the number after the dot starts looking "too big." There is literally no othe
60.
by m3at 4y ago
That's done in professional contexts, when you only care about inference onnxruntime does the job well (including for coreml [1]). I imagine that here apple wants to highlight a more research/interactive use, for example to allow

More ›