MLflow

class sklearn_genetic.mlflow.MLflowConfig(tracking_uri, experiment, run_name, save_models=False, registry_uri=None, tags=None)[source]

Logs each fit of hyperparameters in a running instance of mlflow: https://mlflow.org/

Parameters
tracking_uri: str

Address of local or remote tracking server.

experiment: str

Case sensitive name of an experiment to be activated.

run_name: str

Name of new run (stored as a mlflow.runName tag).

save_models: bool, default=False

If True, it will log the estimator into mlflow artifacts

registry_uri: str, default=None

Address of local or remote model registry server.

tags: dict, default=None

Dictionary of tag_name: String -> value.

create_run(parameters, score, estimator)[source]
Parameters
parameters: dict

A dictionary with the keys as the hyperparameter name and the value as the current value setting

score:

The cross-validation score achieved by the current parameters

estimator: estimator object

The current sklearn estimator that is being fitted