วีดีโอ

Thanks brother, really great content.

Remember, technology transformation is a journey. Organize effectively before diving into the details of Generative AI.

Nice work!

much more useful than the bertopic official document

❤❤❤

Do you have any paper on that topic?

Very informative, thanks

1st like

This is great stuff!

Impressive approach

Plese mention the link of data set

from where we can get the dataset for the study purpose

Informative. Thank you 👍

Thank you very much

The best video ever !!

Nice one mate

In case, if we pass the general data which doesn't contains PHI it is still performing the de-identification of it. How can we solve this challenge ?

ما شاء الله عليكم والله مجهود جبار وعمل عظيم جداً.. جزاكم الله كل خير والله يعطيكم الف عافية👏👏👏👏

Zain is the best in Multi-modal application and weaviate <3 💌

Does she know what is BPE and how to use it?

I'm working on something similar, this is very helpful, thank you for the content!

This is dystopian and insane. Don't worry I don't use anything outside of this platform and sharing anything like this will just get two words invented by the powers that be, said behind my back. In the end you people lose and in the after end up down near the 9th gate. Tik tok, tik tok...counting down. I mean no ill will in my comment. It is what it is at this point. You do it for the money. I am glad I am not working with or for any of these entities in intelligence and government. I have most of my soul intact midway thru life. Even I am willing to admit I was not perfect.

both folks English is to hard to understand

Pls share the presentation deck for all sessions

Is this still supported?

Hey i need your slides please provide me as soon as possible, this is a request. Thank you.

hey can you pleasse give me the slides ?

Good explanation Lawrence. Thanks

good work

Defragment missing :)

Thanks in a million.

Great video. Where can we find the link to the notebooks?

Great content, thanks for sharing!

Awesome.

I am quite confused. If you have to create a JSON table correlating a question with the "measure" and "diagnostic code" your solution just seems rules-based, and the creation of that table would be a herculean task.

is the code open-source? if yes where can i find it? Thanks

AttributeError: 'NLUPipeline' object has no attribute 'predict_embeds' when running FAISS.from_documents(pre_processed_docs, embeddings).as_retriever(), what may be the problem?

Is this code available on GitHub

Very cool

Awesome presentation by Veysel Kokaman. The last leg was the most informative portion which confirmed my understanding that for 'Specific use cases you only need a SMALL model trained using specific data, rather than the 100+MB general purpose LLMs ''. Having your own data is the key for the success.

wow. thanks. Fantastic explanations. This is so practical.

Thank you for this. It was very informative. What are your views on using FactScore?

It's all about drafting proper prompt Cocaman.. anyways thanks for the information, I will try to regenerate these summarisations on the models shown with improvised prompts.

This had Nothing to do with insurance...

Hello Swagata, I really liked the video, thank you. I'm new to the LLM field, but I'm trying to train the GPT-2 model in a specific domain, in this case, an engineering domain. One of the big questions I have is how I should process my data (which are PDF documents) and in what format I should present this information to the model. This is so that after training, I can ask the model questions and it can respond accordingly.

I started researching RAG a week ago and was fortunate to find this video today Thank you from the bottom of my heart for your very rich content

good job👍... Please, can you share the presentation slide ?

Great level of detail in overview and coverage of scale and production-deployment issues (and costs). Many thanks! John Snow Labs does it again.

It's nice to know what a great project we are involved in, congratulations Veysel!!!

Thank you for great presentation. Are you going to share the notebook?