If you want to learn something then try teaching it to someone or else build a #game ๐, especially if you are a #FrontEnd#Developer .
Here's what I built, a simple 2 player game, as a @reactjs@pwa.
https://t.co/2yG8B5L9L2
Big thanks to @PristynCare for an exceptional experience! ๐ From expert doctors to caring staff and top-notch follow-up, everything was perfect. Clean facility, detailed guidance, and genuine compassion throughout. Highly recommended! ๐ #Healthcare#Gratitude
Happiness is the surprise of seeing your son in the parking lot just when you thought you'd missed the chance to give him a goodbye kiss on your way to the office.
#Lemma took Center Stage at @e4mtweets CTV Conference as we proudly unveiled our brand new logo! Our fresh identity reflects our unwavering dedication to pushing the boundaries of programmatic advertising within the ever-evolving world of emerging media formats.
#DOOH#CTV
Open-source Rule-based PDF parser ๐
GitHub - nlmatics/nlm-ingestor: This repo provides the server side code for llmsherpa API to connect. It includes parsers for varioius file formats. https://t.co/BlMtOhZTFj
@kentcdodds Ohh.. this `automagically` scares me... I remember reading it first time in AngularJS docs. Later realised its an alternative nice term for `dirty checking`.
Agreed... we need magical outcomes not magical APIs or components.
Parsing/chunking a document is sooo important to building a prod-grade RAG app, but I havenโt seen a unified library that does both and is tailored for LLM devs..until now! โก๏ธ
Check out LayoutPDFReader (@AmbikaSukla / Kiran), which allows you to parse/chunk PDFs a *context-aware* manner:
โ Hierarchically chunk sections with smaller sections
โ Preserve lists
โ Preserve tables
Full blog post/repo/Colab notebook + examples with @llama_index ๐
Blog post: https://t.co/EfKxlG4wAF
Repo: https://t.co/p5CRXf2HQO
Colab: https://t.co/3lTo2FomQX
Excited to plug this into advanced @llama_index abstractions (e.g. recursive retrieval, auto-merging) to improve RAG performance ๐ช
@mistyHarsh What kind of intelligence are you expecting? Basically, it's generating (spitting out) next chars based on your previous chars. That's what it is trained for. It's very good at predicting.