ProjectConfig#

class scikitplot.mlflow.ProjectConfig(profile='local', session=SessionConfig(tracking_uri=None, public_tracking_uri=None, registry_uri=None, env_file=None, extra_env=None, startup_timeout_s=30.0, ensure_reachable=False, experiment_name=None, create_experiment_if_missing=True, default_run_name=None, default_run_tags=None), server=None, start_server=False)[source]#

Project-level configuration for MLflow usage across multiple scripts.

Attributes:
profilestr

Named profile (e.g., “local”, “remote”, “ci”).

sessionSessionConfig

Session configuration.

serverServerConfig or None

Server configuration (if this profile starts a server).

start_serverbool

Whether to start a managed server for this profile.

Parameters:

Notes

This provides a single, shared configuration for: - train.py - hpo.py - predict.py

It prevents drift between scripts and makes runs reproducible.

profile: str = 'local'#
server: ServerConfig | None = None#
session: SessionConfig = SessionConfig(tracking_uri=None, public_tracking_uri=None, registry_uri=None, env_file=None, extra_env=None, startup_timeout_s=30.0, ensure_reachable=False, experiment_name=None, create_experiment_if_missing=True, default_run_name=None, default_run_tags=None)[source]#
start_server: bool = False#