This project demonstrates a web-based application to query a dataset through natural language using AI-powered solutions.

1. Error Handling - Included try-except block to catch and display errors from both PandasAI and Google GenerativeAI and added voice recognition error handling.
2. Real time Query Suggestions - Added an expander with the sample query suggestions assiting users in formatting queries.
3. Advanced Visualization - Utilized plotly library for plotting the retrieved data which is the actual response.

Travel to the root of the directory !

For this purpose, it uses:

• [Streamlit] to build a data science web app.
• PandasAI to generate Pandas code from a query.
• Google Generative AI (Gemini) to process natural language queries.

Download the dataset "data" folder to the root folder of directory , from dataset.

If you don't have a Python environment available, you can use the conda package manager which comes with the Anaconda distribution to manage a clean Python environment.

Create a new environment and activate it:

conda create -n streamlit-pandasai python=3.9
conda activate streamlit-pandasai

In your active conda environment install dependencies :

pip install streamlit google-generativeai python-dotenv pandas pandasai numpy faker pydantic requests pyyaml SpeechRecognition plotly

1. Get your apis key from Pandasai and GoogleAPI.
2. Replace the 'your_pandasai_api_key' and 'your_google_api_key' with your actual apis key in your main.py file with your actual api key.
3. Save the changes.

streamlit run main.py

Therefore , our query generator not only give the SQL query but also it's explanation , the output and the visual reperesentation of data.