ChatGPT has always been pretty good at generating UML diagrams by providing code blocks containing PlantUML markdown which you can plug into @PlantUML's online interface.
With Code Engineering mode it's even better cos you can pass source code!
Remember you can upload zip files
@YaBoyFathoM Bruh this. It still bamboozles me how disconnected the masses are from new technology. I wish it wasn't the case π’ even though it gives us nerds a head start of sorts π€
Let's hope the future closes the gap on putting revolutionary tech into the hands of the masses quickly!
I made this account to focus on building LLM apps with Python.
I'm not quite ready to share material about that (I feel way behind all the smart people π)
But I'll share some cool colab stuff. I like colab!
This one lets you launch a terminal in colab.
https://t.co/CmS8ezNNiC
@RLanceMartin How come there are so many broken links in the documentation? Is this something that can be solved moving forwards? It seems like pages keep getting moved or the paths modified, but it's a bit of a problem
Document splitting is common for vector storage / retrieval, but useful context can be lost. @langchain has 3 new "context-aware" text splitters that keep metadata about where each split came from. Works for code (py, js) c/o @cristobal_dev, PDFs c/o @CorranMac, and Markdown ..
@pmddomingos If this is the case it's because a lot of folks that try it find it fun, but ultimately see it as a gimmick and don't use it daily.
I'd say it's not because they're moving on to other "amazing" AI tools (that just have GPT backends with custom prompts).
Why do you think it is?
@tunguz It's the same thing that happened when blockchain became hot. Inevitable really. The positive is probably that it will ultimately help separate out the companies really doing something useful with AI. If it's just a label without much substance it'll start to look flaky in time