Quick and easy classification modeling with scikit-learn and pandas
from easyml.modeler import Modeler
modeler = Modeler(df, features, label, model, verbose=True)
modeler.fit()
# Export model
model = modeler.get_model()
# Get metrics
metrics = modeler.get_metrics()
# Save model as pkl file
modeler.to_pickle('model.pkl')
- Quickly iterate with new features
- Experiments (including features, data, and metrics) are pickleable for easy reproducibility
- Basic metrics out of the box
- Track train, test, and full dataset metrics