GRAPE

GRAPE is a regression API in Python environment

Description

GRAPE makes it easy to fit a regression model with hyperparameter optimization. It strings together the workflow of model fitting, hyperparameter tuning, and model diagnostics. (model interpretability coming soon!).

• Available Regression Methods

1. Elastic Net (from sklearn)
2. Random Forest (from sklearn)
3. xgboost
4. lightgbm

• Hyperparameter Optimization
  • Grape Uses Hyperopt's tree parzen estimator

Installation

• pip install grape-model
• If you're having trouble installing some of the dependencies (especially lightgbm, xgboost), try installing them via conda-forge before installing GRAPE

Sample Usage

See sample_grape_use_case.ipynb