And that brings this class on LangChain, Chat with Your Data, to an end. In this class, we've covered how to use LangChain to load data from a variety of document sources using LangChain's 80-plus different document loaders. From there, we split the documents into chunks, and talk about a lot of the nuances that arrive when doing so. After that, we take those chunks, create embeddings for them, and put them into a vector store, showing how that easily enables semantic search. But we also talk about some of the downsides of semantic search, and where it can fail in the certain edge cases that arise. The next thing that we cover is retrieval, maybe my favorite part of the class, where we talk about a lot of new and advanced and really fun retrieval algorithms for overcoming those edge cases. We combine that with LLMs in the next session, where we take those retrieved documents, we take the user question, we pass it to an LLM, and we generate an answer to the original question. But there's still one thing missing, which is the conversational aspect of it. And that's where we finish out the class, by creating a fully functioning end-to-end chatbot over your data. I've really enjoyed teaching this class. I hope you guys have enjoyed taking it. And I want to thank everyone in the open source who has contributed a lot of the things that make this class possible, like all the prompts and a lot of the functionality that you see. As you guys build with LangChain and discover new ways of doing things and new tips and techniques, I hope that you share what you'll learn on Twitter or even open up a PR in LangChain. It's a really fast-moving field, and it's an exciting time. I really look forward to seeing how you apply everything that you learned in this class.
LangChain Chat with your data