Having issue LLM response while performing RAG

I am very new to Lang chain. Watched yoututbe video (https://www.youtube.com/watch?v=yF9kGESAi3M&t=7261s) and am mimicking the codes. While practicing RAG, first error came from TextLoader( "./..//odyssey.txt") - Even when odyssey.txt was present in the directory, it showed "file not found". Solved it by using TextLoader("./..//odyssey.txt", encoding='utf-8').

But when I am using the same query as used in the video- "Who is Odysseus' wife?"

Retriever finds very odd results, no relevancy I can find in the answer. The image shows the retrieved answer by the retriever.

Furthermore, in the case of RAG for conversation, the LLM always generates "answer cannot be found".

I am using Google's embedding - GoogleGenerativeAIEmbeddings(model="models/embedding-001").

I feel something is wrong with vector generation. However, I have also tried with Huggingface embedding model, the issue remains.

Your advice will be helpful.

Thanks for your time.

0 comments

AI Developer Accelerator

skool.com/ai-developer-accelerator

Master AI & software development to build apps and unlock new income streams. Transform ideas into profits. 💡➕🤖➕👨‍💻🟰💰

Leaderboard (30-day)

+23

+21

+15

+15

+13