Hyper-v
Vector databases are so hot right now. WTF are they?
Vector databases are so hot right now. WTF are they?
#Vector #databases #hot #WTF
“Fireship”
Vector databases are rapidly growing in popularity as a way to add long-term memory to LLMs like GPT-4, LLaMDA, and LLaMA. Learn how popular vector databases like Pinecone and Weaviate can store ML embeddings to integrate with tools like ChatGPT.
#programming #tech #TheCodeReport
💬 Chat…
source
To see the full content, share this page by clicking one of the buttons below |
Please invest. This is financial advice: https://github.com/codediodeio/rektor-db
after wathing your video i starded my vector db its 420b if you like send your credit card
Being obsolete twice in one month xD
OMG dude
Most simply a vector has two features: a direction and a magnitude. I think knowing this would help a lot.
Like, embedding is basically grouping of vectors based on similar direction (and possibly magnitude too). Not necessarily distance. And often embedding is in a lower dimension.
such a nice time to be a developer, I just finished reading about ACID and new database paradigm introduced. mind blown
For the love of all that is holy, please, I implore you, KEEP SOME PAUSE BETWEEN YOUR SENTENCES.
Have you compared Graph dbs vs vector dbs?
I'm a little bit ashamed that this fireship guy explains x10 better than me the vector database stuff who have 6 years of experience in ML
8531 5321 4856 1863
06/28 134
Fireship senpai, plis do vector database vid with integration of codellama or wizardcoder into own projects, holy shet I totally missed this video and this looks pretty lit
The idea of vector database seems interesting but you should know there is no such a thing as a relational database since none of the commercial products that sells themselves as relational actually properly implemented the relational data model. A vector database by the way is by definition not a data model.
"Leave your credit card details in the comments if you want to invest"???
AI could learn and memorize autonomously it's frekin crazy. The age of terminator is coming soon
PLZ SLOW DOWN YOUR VIDEOS
Heyyyyy what the fuckkkkk I literally wrote the constant time cosine similarity score algorithm
4905 2185 4418 0973
03/26 325
Still waiting for the Weaviate LangChain tutorial on the second channel.
I have very rarely ever thought of myself as someone's fan. I am yours.
Can you cover Objectbox DB ?
as long as you keep making amazing content like this, you will never become obsolete! 🙂 heart
Oh man how do you put so much effort. Do one video where you showcase how you create these videos, what goes into it, would love to see it
Just trying to think through use cases. Would a user create multiple vector databases for different problems? Embeddings of the same document would differ based on how it was trained, correct? Would embeddings within a use case change over time given more documents?
LLM? not even chat gpt knows what that is….
In higher dimensions all things are close to each other. I think it's time to wipe the dust off of my old Lie Algebra textbooks.
Can't you store vectors in any database? I don't see any difference than me storing it in a good ol sql database and solving these vector similarity re-ranking related questions with the regular tooling such as elasticsearch or faiss