Build your own Streamlit app that lets you TALK TO YOUR DATA with natural language! 🤯

Would you like build a Stramlit app that lets you talk to your data using natural language, whether 500 records or 5MM+ records?
This video will show you how to do this using Streamlit, PandasAI, and a LLM of your choice. Upload large data files and then use natural language to get insights from your data!
Get your PandasAI api key here, you will need it: pandabi.ai
Link to PandasAI documentation: docs.pandas-ai...
Link to the large dataset I used: www.kaggle.com...
Link to many more large datasets on Kaggle: www.kaggle.com...
Link to CODE: github.com/mkt...

[@vasubabujinagam5977]

Amazing tutorial, subscribed 👍

[@varma__123]

Good job

[@schance1666]

Hi Avi, this is exactly what I was looking for except for one thing - is there a way to implement 'rules' that the app must follow? For example, if you have two DBs, one with jobs and one with people, and you want to enforce a rule like: If the job says you must be in Seattle, and the applicant says they only work in Atlanta, that it will not create a 'match'. This has been a problem for me for a long time now (and others I know). If you can do a vid like that, I will share it on all my networks!!! Thanks for the great vid.

[@ahmedelmancy9063]

Hello, thank you for your code, it’s really helpful, I have a problem that the results is NONE, I wonder what I have done wrong, I will submit the picture of the results and tell me kindly what I have done wrong

[@yashk5076]

Hi how do you handle token limit. I have. Built a similar app but using pandas_creat_dataframe_agent.. have a file of 100 columns and 18k record. When I query it the streamlit and terminal shows the result truncated. Also open AI exceeds token limit. Please help me to counter these two issues if u know

[@ahmedelmancy9063]

Sorry I can’t load the picture, the results shows as none word

[@AviPatel68]

Hope you got this figured out. Would love to help, let me know exactly where the code is breaking down.