LocalFileLogger

class LocalFileLogger(experiments_folder: str, config: Optional[Dict[str, Any]] = None, gzip: bool = False)

Bases: etna.loggers.file_logger.BaseFileLogger

Logger for logging files into local folder.

It writes its result into folder like experiments_folder/2021-12-12T12-12-12, where the second part is related to datetime of starting the experiment.

After every start_experiment it creates a new subfolder job_type/group. If some of these two values are None then behaviour is little different and described in start_experiment method.

Create instance of LocalFileLogger.

Parameters

• experiments_folder (str) – path to folder to create experiment in

• config (Optional[Dict[str, Any]]) – a dictionary-like object for saving inputs to your job, like hyperparameters for a model or settings for a data preprocessing job

• gzip (bool) – indicator whether to use compression during saving tables or not

Inherited-members

Methods

finish_experiment(*args, **kwargs)	Finish experiment.
log(msg, **kwargs)	Log any event.
log_backtest_metrics(ts, metrics_df, ...)	Write metrics to logger.
log_backtest_run(metrics, forecast, test)	Backtest metrics from one fold to logger.
start_experiment([job_type, group])	Start experiment within current experiment, it is used for separate different folds during backtest.

start_experiment(job_type: Optional[str] = None, group: Optional[str] = None, *args, **kwargs)

Start experiment within current experiment, it is used for separate different folds during backtest.

As a result, within self.experiment_folder subfolder job_type/group is created.

• If job_type or group isn’t set then only one-level subfolder is created.

• If none of job_type and group is set then experiment logs files into self.experiment_folder.

Parameters

• job_type (Optional[str]) – Specify the type of run, which is useful when you’re grouping runs together into larger experiments using group.

• group (Optional[str]) – Specify a group to organize individual runs into a larger experiment.