aquajet

9 ms·

31.
SpikeGPT: Generative Pre-Trained Language Model with Spiking Neural Networks(arxiv.org)
2 points by aquajet 4y ago|0 comments
32.
by aquajet 4y ago
Yeah, that's one thing that I'm trying to improve. In general it's cool to think of LLMs as a connective tissue for different applications
33.
Show HN: Amazon Product Recommendations Using ChatGPT(getproduct.help)
12 points by aquajet 4y ago|3 comments
34.
Show HN: Using GPT to recommend what Valentine's Day Gift to get(love.getgift.help)
1 points by aquajet 4y ago|0 comments
35.
by aquajet 4y ago
This robot might have a better box jump than me. I want to see a comparison between it and some professional athletes, I think it can seriously compete with them now. I can't wait for the day we have robot (american) football and boxin
36.
Show HN: Using GPT-3 to select holiday gifts(getgift.help)
1 points by aquajet 4y ago|0 comments
37.
by aquajet 4y ago
is the Rust dude on github? I'd like to help
38.
Ask HN: Why did Gmail become popular?
7 points by aquajet 4y ago|26 comments
39.
by aquajet 4y ago
It's entirely possible hacker news/users is a part of the training dataset. It was a part of The Pile ( https://arxiv.org/abs/2101.00027 ), and we included usernames along with the item. Other models have gone
40.
by aquajet 4y ago
Worked on a similar thing here using base GPT3, at least for the first day Replit included so you can verify: https://twitter.com/thiteanish/status/1598217824392351744?t=... I plan on going back and catching up on
41.
by aquajet 4y ago
Are there businesses built explicitly on unbundling? I think there are examples of unconscious unbundling (picking the best individual components of a service/stack) but the goal is never to unbundle. Then a business comes along whose
42.
by aquajet 4y ago
Fixed
43.
Full Stack Startups (2014)(cdixon.org)
40 points by aquajet 4y ago|13 comments
44.
by aquajet 4y ago
Does this still hold true after 8 years?
45.
by aquajet 4y ago
So is the value proposition normally entertainment in these cases?
46.
by aquajet 4y ago
I haven't tried this out personally but have used some of his other tools: https://github.com/anishathalye/periscope
47.
Ask HN: What value does a startup described as a “toy” provide?
1 points by aquajet 4y ago|5 comments
48.
by aquajet 4y ago
How many people use diction? I'm surprised cause I know virtually no one who uses diction, myself included.
49.
by aquajet 4y ago
I'm sure you can set up a raspi to deploy this with minimal running costs.
50.
by aquajet 4y ago
How competitive are the filters? Also, what did you do in the 2 years to make a competitive app? I have published some CS research in the past, so I imagine I would have to do some. Shadowing in addition to the hard constraints of MCAT&#x2F
51.
by aquajet 4y ago
I'm not sure off the top of my head, but it depends on what exactly the specialization is. I'd imagine it's roughly around the salary of a software developer. However, education is a much larger investment, so a lot of that m
52.
by aquajet 4y ago
Thanks, I live in Boston. I know there are EMT courses, so that might work.
53.
by aquajet 4y ago
For context I'm interested in going to med school, but I'm also hesitant because of the long commitment to education.
54.
Ask HN: Has anyone gone to med school after working as a SWE?
9 points by aquajet 4y ago|11 comments
55.
by aquajet 4y ago
I've been experimenting with a search engine/personal assistant to index the long list of material for me, and feed back snippets/articles as I want. I guess the analogy is having a person "read" the material for me
56.
by aquajet 4y ago
> Training a sequence-to-sequence language model (T5 derivative) on our custom dataset designed for factual generation yielded much better results with less hallucination. Could you elaborate more on this or point to a paper/benchma
57.
by aquajet 4y ago
How do you get it to work with sites that are javascript-heavy or have heavy bot restrictions? I've been trying to make something similar but spent weeks just fighting edge cases on HTML not caching correctly.
58.
by aquajet 4y ago
I've been using upvotes as a way to save things, and then using a script ( https://github.com/anishthite/HN-Saved-Links-Export ) to export them to json so I can search through them.
59.
by aquajet 4y ago
Why so much switching?
60.
by aquajet 4y ago
I've been working on a solution to this: https://diva.so mainly cause I had the same issue. It's a third-party service unfortunately, but it can index the contents of your bookmarks + other sources to let you search th

More ›