Financial Report Retrieval Augmented Generation

Objective:

There can be multiple similar industry stock that are interesting to analyst but time consuming to go through each of the report. This repo is using current RAG method to index in one or more company and make comparison between two or more different companies.

Stack in use

• llama-parse for document parsing
• GPT4o-mini for query and embedding
• langsmith for observability

STEP 1 : Start

To start with :
edit .env with llamacloud and openAI API key.
llamacloud : sign up at https://cloud.llamaindex.ai > API Key > Generate New Key
openAI : sign up and go to https://platform.openai.com/api-keys > create new secret key
langchain : sign up and go to https://smith.langchain.com/settings -> API keys


Insert all key into .env

STEP 2: Report parser

run through notebook/financial_report.ipynb

Suggestion of what to add and enhance

Please log functionalities that you wish to add or enhance in issues section.

Contributor are all welcome