Curated Resource ( ? )

A step-by-step guide to building a chatbot based on your own documents with GPT

my notes ( ? )

" a step-by-step guide for building a document Q&A chatbot in an efficient way with llama-index and GPT API... ask the bot in natural language about your own documents/data... [see it] retrieving info from the documents and generating a response [1]... customer support, synthesizing user research, your personal knowledge management"

Key points:

  • "fine-tuning costs quite some money and requires a big dataset ... also impossible to fine-tune every time when there is a change to the document.... so for (multi-)document QA [Question Answering], fine-tuning is not the way to go.
  • alternatively, prompt engineering: "append the original document content before the actual question... [but] it can only take in a few thousand words ... costly... because the pricing is based on the number of tokens you use.
  • hence this approach: "using an algorithm to search the documents and pick out the relevant excerpts and then passing only these relevant contexts to the GPT model with my questions...
  • LlamaIndex does exactly what I wanted to do and it is straightforward... convert your original document data into a vectorized index... use this index to find the most relevant parts based on the similarity of the query and the data... plug in what’s retrieved into the prompt it will send to GPT [for]... context"

Read the Full Post

The above notes were curated from the full post bootcamp.uxdesign.cc/a-step-by-step-guide-to-building-a-chatbot-based-on-your-own-documents-with-gpt-2d550534eea5.

Related reading

More Stuff I Like

More Stuff tagged knowledge management , guide , chatgpt , llm , llama

Cookies disclaimer

MyHub.ai saves very few cookies onto your device: we need some to monitor site traffic using Google Analytics, while another protects you from a cross-site request forgeries. Nevertheless, you can disable the usage of cookies by changing the settings of your browser. By browsing our website without changing the browser settings, you grant us permission to store that information on your device. More details in our Privacy Policy.