Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
rg111
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
46 ms
·
511.
▲
Smartphones will kill off the DSLR within three years says Sony
(techradar.com)
3 points
by
rg111
4y ago
|
0 comments
512.
▲
by
rg111
4y ago
Social audio is not dead. It existed for decades, and it was known as... conference call. Before internet, we used conference calls in our friends circle and among relatives, a lot. Now we do a group call in Meet or Telegram. And Discord h
513.
▲
by
rg111
4y ago
I have actually done both, and the time taken to do the same task in PowerPoint is much lower. I would say 1/5th or 2/5th. Also, as I said, the experience of writing code with Manim Community is not good. If you have experience us
514.
▲
by
rg111
4y ago
I have made animations with it with enough Math and Deep Learning that were presented both in-company and outside where work done with animations made in PP even won awards. The Deep Learning papers one sees in NeurIPS, ICML, CVPR, etc. alm
515.
▲
by
rg111
4y ago
I have transitioned to PowerPoint. I create animations and then export a video file.
516.
▲
by
rg111
4y ago
Manim, as maintained by Grant Sanderson has many rough edges and he discoirages its use by other people. There's a more stable downstream known as Manim Community. One can use it, it has somewhat good documentation. But even this leads
517.
▲
by
rg111
4y ago
What the OP and people in all replies that I read really misses one thing: Rejecting AI-products might be an option for you, but it might not be an option for everyone. Not everyone will be able to. Think of coal. Or car and factory emissio
518.
▲
by
rg111
4y ago
Can I please have a link?
519.
▲
by
rg111
4y ago
I use code search in You just for the layout and the variety of resources I see. On the other hand, Kagi can show me better results that are better than Google. You often includes many results in Google search results. Kagi has results that
520.
▲
by
rg111
4y ago
> But let's be clear, the fact that Italy - my country - has been fragmented more than it was unified doesn't justify the argument that Italians haven't existed for more than 2 thousand years. Totally agree with that. My f
521.
▲
by
rg111
4y ago
> they've been a unified entity for over 3 thousand years If you look at China's history a little bit deeply, you will notice a rhythmic pattern of unification and fragmentation. I know this because I was told a by friend who i
522.
▲
by
rg111
4y ago
Google search gives me mostly shit results now. I have now completely switched to Kagi and code.you.com. I barely use Google now.
523.
▲
by
rg111
4y ago
In my opinion he is the only comedian worth tuning in to right now. I just love the man.
524.
▲
by
rg111
4y ago
Yrrk. Transformers are being used in Deep RL for at least months. Try these: https://scholar.google.com/scholar?q=transformer+deep+reinfo...
525.
▲
by
rg111
4y ago
Nice resource, but still 10+ hours of video and nothing else. No code, coding assignments, math problems or coding problems. Very little RoI. I watched them all from start to finish. I had a superficial, shallow "understanding" bu
526.
▲
by
rg111
4y ago
Exactly. I got perfect scores in both of the Mathematical Physics papers in college. Many problems, with some variations did appear directly from Spiegel. Best VC problems book ever.
527.
▲
by
rg111
4y ago
By that book, do you mean the one authored by Spiegel? I went over that book fully. From cover to cover. It was absolutely fantastic, and as good as a "problems book" can be. I loved it and would recommend it to anyone who is look
528.
▲
by
rg111
4y ago
Google Duo has the best video quality in video chats in Android. This has been my first-hand experience.
529.
▲
by
rg111
4y ago
1. Nature of Code - simulating and modeling many things in nature. 2. Data Science from Scratch - building a Data Science library from scratch. 3. Deep Learning from Scratch - building a Deep Learning library from scratch. 4. Realm of R
530.
▲
Deep Reinforcement Learning Course from Hugging Face
(github.com)
7 points
by
rg111
4y ago
|
0 comments
531.
▲
Code, Weights for VQ Diffusion Model for Text-to-Image Sythesis from Microsoft
(github.com)
13 points
by
rg111
4y ago
|
1 comments
532.
▲
by
rg111
4y ago
You are yet to get your first job. Your first and immediate goal should be doing projects. Pick a popular website and clone it with Django/Flask. Do it for 2/3 websites. Write and maintain a blog post where you share your journey
533.
▲
by
rg111
4y ago
This is the only sensible thing to do when uprooting your family and the changing the course of life of multiple people.
534.
▲
by
rg111
4y ago
@Perez418, this person can give you better advice than I. I have been in a tech since the last year of college. This person has more relatable career trajectory as yours.
535.
▲
by
rg111
4y ago
Offside is a lot easier to understand than the whole rules of Baseball.
536.
▲
by
rg111
4y ago
Feynman Lectures are for people who already have good footing in the topics that are being covered there.
537.
▲
by
rg111
4y ago
I wouldn’t really recommend Berkeley Physics series to self-learners. Without a teacher's help, it would be really difficult. I don't think it is impossible.
538.
▲
by
rg111
4y ago
Physics Nobel Laureate Gerard 't Hooft has put together a comprehensive list of resources for people wanting to learn: How to become a GOOD Theoretical Physicist [0]. You should check it out. [0]: https://www.goodtheorist.sc
539.
▲
by
rg111
4y ago
I have been a Physics nut for majority of my life (not any more, though). My question is- what do you want to do with your Physics knowledge? What’s your goal? Just intellectual curiosity, or something else? Please state that. Is there an
540.
▲
by
rg111
4y ago
Send me something, like a Discord username, or an email. Best would be Libera.chat, though. Update your profile, and I will check back after 24 hours.
More ›