Video

Outstanding well done!

Excellent content, however if you speak slowly it will make it easy to absorb the information

thank you ryan gosling

"Filler words" this is hugely important, IMO. Because it's literally how human beings speak. They use these delaying tactics constantly.

Simple & Clear explanation.

Loved your channel, could you do one with LangServe, please, thanks.

.unique() is not working for everything,while datacleaning some extra information it was showing df['TypeofContact'].unique()

Hi Greg. Great video! Thanks for sharing. But I have some issues when running the code: Firstly, the Speech-to-text performance is not very good. I literally have to shout into my mic for it to be able to hear. I thought it was my microphone issue and tested it normally on a simple voice recorder and it worked as expected. Secondly, the text-to-speech voice output keeps breaking in between. Not sure if that is expected because of ffplay but definitely wasn't as smooth as what you showed in the video. Thirdly, the voice input is not getting recognized immediately once I get the response. It seems there is a small but noticeable delay to when I get the response back as voice to when I can again start speaking even though it says "Listening" in the console. I have to wait for like 5-10 seconds before I start speaking for the program to recognize my voice or else it isn't doing so. Is anyone else facing the same issue?

Great video! so what if my question is out of context of the pdf document? Will the open ai answers it from its generic knowledge? Or it will simply say that it doesn't know the answer? Either way can we configure it to respond the way we want?

I guess semantic search is what already vector databases are supporting.. and i don't find any advantages by doing sentence split, then calculating cosine distance and putting them in same bucket.. am i missing something here?

Single and multi hop explained concerning the semantic splitting. Nice !!

Iterrows is NOT faster than itertuples.

Not a comment on the context, but I think that the style of the thumbnail is very smart. It reminds me of the Wired 5 Levels of Difficulty style. I think if the aesthetic is softer, it can be even more popular.

group rank is really intuitive and simple, but I don't really understand how normalizing ranks works? and what's the idea of applying it?

The voices are all terrible, imho. They would work well in a cyberpunk game - but that's about it. :D

Amazing video as always, thanks Greg. Did anyone had some issues with ffplay?

thank you

I can't appreciate this video or this playlist more. This work is a masterpiece. Thank you!!

man it took me 3 weeks to find you. thank you please keep on coming.

I've a better way for imsgesyin PDF, I converted PDF to markdown and turn all images to markdown image reference, then put all the images in a separate folder, that way my embeddings show markdown text with markdown images which can later be turned into html images in the chatbot

Thanks! you helped me to get to the right path to my solution !

Thank you. I could not get my code to run. Thank you….

Very helpful Video, Thank you!

I can see video was posted a year ago. There is one better approach, which I'm not sure was available year ago. You can use RAG + Data Embedings. Each document chunk you can convert to vectors and store in vector store/database. Then also prompt can be converted into vector and matched against vector store/database, which will give you closest match. Then just use GPT model on best match found earlier.

can somebody helps me with interruption handling in this provided video as I have created a voicebot which is running on UI but unfortunately it cannot handle voice interruptions.

LOL I think this entire video is made with an A.I. clone of you :P

Really Informative 😊

I have a use case with a haven't seen anywhere: I create a private GPT that has documents as contexts. This documents has criteria for a specific subject I give it system instructions so its function is to evaluate a user document that it's attached as a part of the prompt, to see if the document complies with the criteria in the context documents, and give detail response on the result of evaluation and the justification that includes, the content of the user document and the criteria in the context documents. I want to do that in LangChain but I don't know hot to add a user document as a part of the prompt for the RAG.. It would be great if you can you explain how to approach this implementation. Thanks you for the content!! Keep the good work.

pricing, pricing, pricing always pricing.

Great Insights

Hey! Does the 4o version use something similar?

Rip to semantic search for large datasets 😂! But interesting approach.

Love how straight to the point the explanation was

8:52 Thou shan't troll the beginners for you were one too. - Pro Code, Rule 4.

Nicely done

This is an amazing professional content! it hits the point directly

This is an insanely detailed from first principles tutorial. Thank you for taking the time to put this together.

While running the code elements=partition_pdf(filename=filename,strategy="hi_res",infer_table_structure=True,model_name="yolox") in my Jupyter notebook, I encountered errors such as TesseractNotFoundError. If anyone has faced this issue or knows how to solve it, please guide me as soon as possible. Great work, sir. I can confidently say that such a combination of content and explanation is unparalleled in the CZcams world. @DataIndependent

My question is pinecone only stores vectors and not text files, how do I get the texts in my program

Love this! I'm working with transcripts where semantically generated chunks can be quite large. These chunks need to be further divided to fit the limits of the embedding model. Given this, isn't semantic chunking unnecessary if we ultimately have to recursively break down the larger chunks into smaller ones?

You Good sir are not a Guy who talks the talk. I learnt more about Langchain from you in this half hour than anybody else I have listened to on the last 3months. The veritable quarter to be exact. You are a Guy who walks the walk 🫡

Liked this semantic splitting! Cool stuff you´ve done there!! Also agentic chunking. Pretty cool!!!

great video, thanks you

Hey Greg, thanks for this video! Since, there is a limit to access open ai api key without paying, how can the above implementation be carried out with other open source LLMs ?

Fantastic tutorial! It would be great to see another tutorial using "transformers" instead of openai with chroma or any local database... and how will you save the extracted information.. does Kor tokenize that information, etc?

Great video! it helped me clarify the past and present of all the chunks. I have a question, in agent chunking, there can be an issue of having too much content on a single topic. In extreme cases, an entire book might be about one topic. How should we further break it down in such situations?

That was really helpful, thank you for information you're sharing!

was working then: RuntimeError: Task <Task pending name='Task-10' coro=<AsyncLiveClient.send() running at C:\Users\Sarah\.conda\envs\deepgram\lib\site-packages\deepgram\clients\live\v1\async_client.py:251> cb=[_chain_future.<locals>._call_set_state() at C:\Users\Sarah\.conda\envs\deepgram\lib\asyncio\futures.py:392]> got Future <Future pending> attached to a different loop