Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/concaption/streamlit-langchain-csv-qna
A streamlit based chatbot for custom CSV data.
https://github.com/concaption/streamlit-langchain-csv-qna
chatbot csv langchain python streamlit
Last synced: 4 days ago
JSON representation
A streamlit based chatbot for custom CSV data.
- Host: GitHub
- URL: https://github.com/concaption/streamlit-langchain-csv-qna
- Owner: concaption
- Created: 2023-09-08T23:30:37.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2023-09-09T03:12:20.000Z (over 1 year ago)
- Last Synced: 2024-12-16T02:12:18.906Z (about 2 months ago)
- Topics: chatbot, csv, langchain, python, streamlit
- Language: Python
- Homepage: https://langchain-csv-qna.streamlit.app/
- Size: 184 KB
- Stars: 3
- Watchers: 1
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Streamlit CSV Q&A Chatbot
[](LICENSE)
[](https://www.python.org/downloads/)
[](https://pypi.org/project/your-package-name/)
[](https://github.com/concaption/streamlit-langchain-csv-qna/graphs/contributors)
[](https://github.com/psf/black)
[](https://github.com/concaption/streamlit-langchain-csv-qna/issues)
[](https://github.com/concaption/streamlit-langchain-csv-qna/pulls)
## Overview
This is a chatbot application built using Streamlit and a series of NLP tools. The application allows users to upload a CSV file containing textual information. The chatbot can then respond to questions about the data stored in the uploaded CSV file.
## Todo
- [ ] Write a blog post on it.
- [ ] Refactor
- [ ] Re Write code such that it can be used a tutorial.
## Installation and Usage
### Prerequisites
- Python 3.6 or higher
- Git
- OpenAI API Key
### Clone the Repository
To clone the repository, run the following command in your terminal:
```bash
git clone https://github.com/concaption/streamlit-langchain-csv-qna.git
cd streamlit-langchain-csv-qna
```
### Create Virtual Environment
```bash
python -m venv .venv
source .venv/bin/activate
```
### API Key Setup
Create an OpenAI API key from [here](https://platform.openai.com/account/api-keys) and add it to `.streamlit/secrets.toml`:
```toml
open_ai_api_key = "your-api-key-here"
```
### Install Dependencies
Run the following command to install the necessary packages:
```bash
make install
```
### Linting and Formatting
To lint the code, run:
```bash
make lint
```
To format the code, run:
```bash
make format
```
### Testing
Run the following command to execute tests:
```bash
make test
```
### Running the App
To run the application, execute:
```bash
make run
```
## Features
- Upload a CSV file to query against
- Ask questions in natural language about the uploaded CSV data
- Uses OpenAI's GPT model to generate human-like responses
- Session storage for previous interactions
- Secure OpenAI API key storage via Streamlit secrets
## Workflow Commands
- `make install`: Install dependencies
- `make lint`: Lint the code
- `make format`: Format the code
- `make test`: Run the test suite
- `make run`: Run the application
- `make refactor`: Run lint and format
- `make all`: Perform install, lint, format, test, and run in one command
## Architecture
The architecture is fairly straightforward:
- `CSVLoader`: For loading and handling the CSV file.
- `VectorstoreIndexCreator`: For indexing the loaded documents.
- `RetrievalQA`: For retrieving answers from the indexed documents.
- `OpenAI`: For leveraging OpenAI's GPT model to generate natural language responses.
## License
This project is licensed under the MIT License - see the [LICENSE.md](LICENSE.md) file for details.
---
Made with :heart: by [Usama Navid](buildberg.co/about). For any issues or suggestions, open an issue.