StarMorph AI
Introduction to Langchain Javascript Documentation. How to Create GPT-3 GPT-4 Chatbots that can contextually reference your data (txt, JSON, webpages, PDF) with embeddings . Brief Introduction into embeddings, vectorstorage options such as Pinecone, Chroma.
Intro Call https://cal.com/starmorphai/intro-call
1 hr consulting https://cal.com/starmorphai/consultingcall
website https://starmorph.com
Video Resources
NextJS Trained Bot: https://code.chat
Langchain JS Docs: https://js.langchain.com/docs/
Langchain Python Docs: https://langchain.readthedocs.io/en/latest/
OpenAI Embeddings Docs: https://platform.openai.com/docs/guides/embeddings/use-cases
Zahid Open Source Langchain: https://github.com/zahidkhawaja/langchain-chat-nextjs
I have updated the title, description, and thumbnail to remove the incorrect description of training. This strategy allows you to contextually reference your data with embeddings, not train the underlying model
When follow up video?
fantastic video exactly what I needed
Hey, just saw this video, loved how you explained the concept, but I was actually expecting a small example of how to query/converse with the data once you generate the vector
could you make a tutorial more in-depth about the code for the chatbot
Now imagine training GPT with the whole body of your emails
Hey man! Good work on the video! I have been in the planning stages of a really cool project with this kind of tech. If you wanna collaborate, I would be down. Let me know. Do you have a discord? Would love to connect and talk.
i clould not install hnswlib-node doest it need to install vscode build tool with desktop envirment c++ ?
Hey, what do you use to get data from a website? Puppeteer?
Would you kindly consider removing all noise such as you typing and the annoying bell asking for engagement? ❤ Please consider keeping only your voice.
got a link to your repo?
Do you do private projects for hire?
Can you please make a video of making this project from scratch.
Great video! Just wondering how you managed to achieve the streaming of answers; I'm working on something similar on AWS Lambda, but I can't get the streaming to work
Does this keep my data private? I don't want my data shared with the entire LLM. Thanks
Hey I'm really interested in having a conversation further about Vector embeddings. I am working on a school project and would greatly appreciate your knowledge to help me out
My man! I’ve been watching lot of videos about langchain, and as a beginner programmer, the simplicity of this video was so, so great. Not the “most value”, but definitely the best first video to watch on the topic
Thanks for this very helpful video. I am very happy to have found it. I would like to create my own knowledge database that includes ideas I've had in the past, transcripts from YouTube tutorials, manuals, etc. Some of this content is constantly changing. For example, a manual isn't updated very often, so I can create a database with it and leave it as is. However, what about ideas and the aforementioned transcripts of tutorials? It might not be very efficient to manually update these, so I would try to automatically update them on a schedule. Alternatively, would it make more sense to leave them out and just let the bot scan through all my ideas in normal text form, notifying me if I've had a similar idea before or if it finds connections between different ideas I've had? Any help is appreciated. Thank you.
Greate video , I wanna to know how much does it cost for training Next13 js ?
are there more tutorial on langchain? i just pip installed it, not knowing how to start trainning on my own data
Please give me your email, i want to do a project with you
Great vid man 🎉❤ would be awesome if you did an e2e video of the nextjs docs chat you displayed 😘
I have a question, once i train the chatgpt, obviously the answer of the new trained model will be more accurate. But will chatgpt use my data publicly ? Or this trained chatbot is a private account ? I am asking since I may use confidential data..
Great Video, I was exactly looking for an Explanation like this. Helped me a lot. thanks buddy!
What are the best practices for ingesting large files? How much data can you store using the in-memory dbs (e.g. Chromadb, HNSWLib()? I'm trying to generate embeddings for a large CSV (100k rows).
Can someone assist me withba good programed that can assit me set up
Can I use my own data that contains bad words and offensive language?
What is the difference between referencing your data versus training the underlying model? I'm assuming referencing means it can ingest patterns from your own data and create novel content informed by that specific data – or does it just spit out bits and pieces of your own data?
Train a bot from start to finish, please! You are teaching a 61 year old guy a new thing. Thank you.
Imagine train a bot on MDN
Bro! the subscribe audio is too loud almost break my ears
We should work on a project together! If you see this and are interested, let me know. I can do front and backend.
Is it free or you have to use the same OpenAI api's key used for the api integration for example in a website?
Does langchain have an chat bot I can talk to that is trained on the langchain docs….?
hi , is it possible to train gpt model on video format data ?
i mean programing tutorial video so it should analyze the images and the codes that the instructor is explaining on video,
is it possible with langchain and javascript ?
Thanks for the content, really helpful!
I wanted to ask you, are these embeddings kind of expensive when we talk about OpenAI Tokens, aren't they?
And is it true that currently there are still no ways to properly train a custom ChatGPT model?