The objective of this project is to develop a Virtual Assistant Bot on Whatsapp that can receive and process diverse types of documents, including research papers, enabling users to obtain accurate and timely responses to their inquiries.
In order to train the model on the submitted document, I utilized the OpenAI Embeddings API. This involved constructing a dataframe from the document and assigning tokens to track the length of sentences. Using the embeddings API, I generated embeddings for the dataframe, which can subsequently be used to match user questions and provide accurate responses.
I am encountering a RateLimitError due to the utilization of the free API for this project, which is subject to certain restrictions. However, opting for the paid API would eliminate this issue entirely.