minimaxir/automl-gs

Health v2 · maintenance only

Adoption not part of the score

1866 stars · 178 forks observed · 2026-09-21

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

automl-gs is a Python AutoML tool that takes an input CSV and a target prediction field and automatically generates a trained machine learning or deep learning model along with native Python pipeline code. It abstracts data transformation, model architecture, and hyperparameter search across frameworks like TensorFlow/Keras and XGBoost with no black-box lock-in.

Use cases

• predict a target column from a csv without writing model code
• generate a baseline model for tabular data automatically
• get native python code for a trained prediction pipeline
• run automated hyperparameter search on a dataset
• build ml models without deep statistical knowledge
• export a model pipeline with no platform lock-in

When to choose

• you have tabular csv data and want a strong baseline model quickly
• you want readable, editable generated code instead of a black-box model
• you're a citizen data scientist without deep ML engineering experience
• you want to train models cheaply using TPUs in Google Colaboratory

When to avoid

• you need state-of-the-art model performance beyond strong baselines
• you need actively maintained tooling with recent updates
• your data is unstructured like images, audio, or free text
• you need a full AutoML platform with extensive framework support

Facets

library · maturity maintenance

machine-learning deep-learning data-science etl machine-learning data-science python cli cross-platform automl tabular-data hyperparameter-search code-generation tensorflow xgboost citizen-data-science automation

Member repositories

For agents

markdown · JSON · MCP: product_card(name="minimaxir/automl-gs")