I built an open-source library to create Machine Learning models using natural language

I've built smolmodels, a fully open-source library that generates ML models for specific tasks from natural language descriptions of the problem. It combines graph search and LLM code generation to try to find and train as good a model as possible for the given problem while experimenting with various model architectures. Here’s the repo: https://github.com/plexe-ai/smolmodels Here’s a stupidly simplistic time-series prediction example: import smolmodels as sm model = sm.Model( intent="Predict the number of international air passengers (in thousands) in a given month, based on historical time series data.", input_schema={"Month": str}, output_schema={"Passengers": int} ) model.build(dataset=df, provider="openai/gpt-4o") prediction = model.predict({"Month": "2019-01"}) sm.models.save_model(model, "air_passengers") The library is fully open-source, so feel free to use it however you like. Or just tear it apart in the comments if you think this is dumb. I’d love to get some feedback, and the project is very open to code contributions!