Caution
This project doesn't work properly, yet.
Notebook Data Scientist (nbds) is a PoC project to analyze data by Large Language Model (LLM) on Jupyter Notebook.
This project was inspired by Data Interpreter.
from nbds.models import Gemini
from nbds import NBDataScientist
api_key = ...
model = Gemini(api_key)
scientist = NBDataScientist(model)
scientist.register_magic()%%nbds
Read "Training Data" and create a model to predict "median_income",
then evaluate the model with "Test Data"
Training Data: ./sample_data/california_housing.csv
Test Data: ./sample_data/california_housing.csv
Warning
We execute LLM-generated code. Every time the code will be displayed and you will be asked before execution. You must check generated code by yourself. Additionally, we highly recommend to run separated environment like container, and not to analyze any confidential information at all.