Activity
Mon
Wed
Fri
Sun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
What is this?
Less
More

Memberships

Data Alchemy

35.2k members • Free

188 contributions to Data Alchemy
I want to build something with AI that looks at my photo album
Hello! Do you guys have any recommendations on how I can go about this? It is a vague idea I know, but I thought it would be fun to do if I can use AI to categorize them or something.
1 like • 1d
Google Photos does some kind of stuff like this if you have them uploaded there. If you prefer something more local and less of a prying eye, maybe you can look at github or some other online repository. I think that recognizing pictures is one of the AI weaknesses as the captcha still uses images to tell humans from bots, but I can refer to the first tool to say that does impressive stuff uniting pictures from the same place but from different angle and making one collage or poster. Let's see if there is anyone else that can give you a better advice. Maybe @Pierre-Henry Isidor . He's got several great ideas and projects and maybe one fits yours.
0 likes • 11h
@Pierre-Henry Isidor Awesome! Thank you. 😎
1 like • 15h
@Boon How Ho Great question. A lie is, per definition, something said to be true knowing it isn't. So, how does the agent know something isn't true? Sometimes even humans can't tell it apart...
Need some help leveling up 🙏
Please like and comment I will help you back for sure 🙃
1 like • 3d
@Marcus Zographos I am at this stage too. I have done some Space Invaders clone with Python following the instructions of Eric Matthes' book "Crash Course on Python" (see more details here: https://www.skool.com/data-alchemy/data-alchemy-book-club?p=12996f1d), but no more. If you prefer some other manual, or tutorials on internet, there are plenty of examples. I'd recommend you to follow the YT channel Programming with Mosh (https://www.youtube.com/@programmingwithmosh). It has simple, pretty straightforward and very understandable videos covering a wide arrange of topics, including Python. Dave Ebbelaar, the man behind this site, has also a YT channel (https://www.youtube.com/@daveebbelaar) where you can find a wide variety of tutorials and instructional videos. Both explain the topics without the hype or trying to sell you something. Both have businesses running and will tell you about them, but you can follow their videos without having to subscribe or pay for stuff. But if you read some interesting article on the internet, see a video worth sharing or have a question to ask, you can post it here. Also, you can share your thoughts on what others post and comment. These ways you'll get likes and help you progress leveling up. I hope that helps.
2 likes • 2d
@Boon How Ho Ok, here's some advice: If you want to level up you have to contribute with quality content. You can talk about your experience, ask questions, answer topics, etc. Read the pinned posts, where the basics are explained, and you will know how this place goes. This said, I welcome you here and hope you have a great experience, learn a lot and enjoy your time here.
Building AI Agents
Anthropic has published a very interesting article about building AI agents. Have you read it? Anthropic engineers suggest that you should understand the underlying code before using any abstraction frameworks like LangGraph; otherwise, it may get hard for you to debug and understand your own code. Also, they suggest that you should only add complexity when it's really necessary for your use case. Simplicity is the best, and you shouldn't build agents when you don't need them. They also provided a detailed overview of commonly used agentic systems. Have a look at it 👉 https://www.anthropic.com/engineering/building-effective-agents
2 likes • 3d
Yep. Seen and read it. I think it's a very interesting approach and gives good advice. What in a not-so-far future may distance winners from losers is the ability to understand what they do and explain it to whoever asks it. Also, as an old adage says "Less is more". Simplicity is the key to do things, the key to explain it, and the key to easy upgrades. Alongside with organization, those are -in my opinion- the core blocks of a good work.
2 likes • 3d
@Istiaque Reza Auyon You're welcome. Hope it helps.
Godfather of AI has a bold new plan to keep us safe from it.
I posted this link as a reply in another conversation, but now think it is worthy of its own. It's very interesting and eye opening, especially the Q&A. While Yoshua Bengio discusses unintended consequences of AI and corporate profit drivers, I wish he would have touched intentional unethical and malicious implementation of AI, we have already seen some examples. These users are unlikely to adhere to safeguards. https://www.vox.com/future-perfect/417087/ai-safety-yoshua-bengio-lawzero
Godfather of AI has a bold new plan to keep us safe from it.
1 like • 3d
@Nikolaj Genov I see your concerns and understand them. I think that one thing will be the military-AI and another one the civilian-AI. AS per now there is a lot of technology used in the military that is not available for public use and I doubt it will ever be. Think of one thing as stupid (?) as helicopters. Civilian ones are completely different from military ones despite the main operating principles are the same. The Indian army is using AI on their attacks on Pakistan and their defenses on counterstrikes from the Pakistani army. Sooner or later that will be a reality in many armies worldwide and will happen something like it did with atomic bombs: lots of countries developed it to avoid falling behind. So civilan-AI can be restricted using the LawZero principles because the main ones are made by companies, not non-profit organizations, and they will -sooner or later- realize they can't allow the general public acces such a powerful unleashed tool as an "unethical" or unrestricted AGI or ASI. We, the working class, will have access to an "ethical" -or Asimovian- version of the AI/AGI/ASI. Be political, as technological progress is. And someone said that if you don't do politics, someone else will do it for you. Politics aren't bad. Being a bigot is.
1-10 of 188
Oriol Fort
6
1,296points to level up
@oriol-fort-2227
Former archaeologist turned into teacher. Learning about AI to be able to create programs that make a difference for students and researchers.

Active 11h ago
Joined Jan 20, 2024
powered by