Welcome to this repository! This Ask-CSV-Agent repository contains a project that leverages the capabilities of chatGPT to answer questions based on a database of Nike products descriptions.
In this project, I utilized data describing Nike products to create a chat agent capable of answering questions related to those products. For instance, I asked my chat agent, "What is the best shoe for football?" and received the following response, based on the database I provided: "The Nike Dri-FIT shorts are the best shoes for football. They are designed to wick sweat away from your skin while you run, dribble, and shoot." This functionality is akin to having a knowledgeable sales associate at Nike who can answer questions tirelessly.
To run this project, you'll need to install the following dependencies:
- tabulate
- faiss-cpu
- langchain
- openai
- (You can also use another model, such as PaLm 2)
Additionally, you'll need to access the OpenAI API. You can find more information on how to obtain API access on the OpenAI website.
I retrieved the data used in this project from Kaggle. The dataset contains Nike product descriptions that were used to train the chat agent.
To interact with the chat agent and query the database, you can refer to the Ask-DataBase.ipynb
file included in this repository. It provides three different methods for accessing and exploring the database.
If you have any questions or would like to get in touch with me, you can reach me on
- Facebook: Mustafa Mohammad
- Telegram: Mustafa Mohammad
Thank you for checking out this project!