etna.loggers.WandbLogger#

class WandbLogger(name: str | None = None, entity: str | None = None, project: str | None = None, job_type: str | None = None, group: str | None = None, tags: List[str] | None = None, plot: bool = True, table: bool = True, name_prefix: str = '', config: Dict[str, Any] | None = None, log_model: bool = False)[source]#

Bases: BaseLogger

Weights&Biases logger.

Note

This logger requires wandb extension to be installed. Read more about this at installation page.

Create instance of WandbLogger.

Parameters:

  • name (str | None) – Wandb run name.

  • entity (str | None) – An entity is a username or team name where you’re sending runs.

  • project (str | None) – The name of the project where you’re sending the new run

  • job_type (str | None) – Specify the type of run, which is useful when you’re grouping runs together into larger experiments using group.

  • group (str | None) – Specify a group to organize individual runs into a larger experiment.

  • tags (List[str] | None) – A list of strings, which will populate the list of tags on this run in the UI.

  • plot (bool) – Indicator for making and sending plots.

  • table (bool) – Indicator for making and sending tables.

  • name_prefix (str) – Prefix for the name field.

  • config (Dict[str, Any] | None) – This sets wandb.config, a dictionary-like object for saving inputs to your job, like hyperparameters for a model or settings for a data preprocessing job.

  • log_model (bool) –

    Log checkpoints created by pytorch_lightning.callbacks.ModelCheckpoint as W&B artifacts. latest and best aliases are automatically set.

    • if log_model == 'all', checkpoints are logged during training.

    • if log_model == True, checkpoints are logged at the end of training, except when pytorch_lightning.callbacks.ModelCheckpoint.save_top_k==-1 which also logs every checkpoint during training.

    • if log_model == False (default), no checkpoint is logged.

Methods

finish_experiment()

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.

reinit_experiment()

Reinit experiment.

set_params(**params)

Return new object instance with modified parameters.

start_experiment([job_type, group])

Start experiment.

to_dict()

Collect all information about etna object in dict.

Attributes

This class stores its __init__ parameters as attributes.

experiment

Init experiment.

pl_logger

Pytorch lightning loggers.
finish_experiment()[source]#

Finish experiment.

log(msg: str | Dict[str, Any], **kwargs)[source]#

Log any event.

e.g. “Fitted segment segment_name” to stderr output.

Parameters:

  • msg (str | Dict[str, Any]) – Message or dict to log

  • kwargs – Parameters for changing additional info in log message

Notes

We log dictionary to wandb only.

log_backtest_metrics(ts: TSDataset, metrics_df: DataFrame, forecast_df: DataFrame, fold_info_df: DataFrame)[source]#

Write metrics to logger.

Parameters:

  • ts (TSDataset) – TSDataset to with backtest data

  • metrics_df (DataFrame) – Dataframe produced with etna.pipeline.Pipeline._get_backtest_metrics()

  • forecast_df (DataFrame) – Forecast from backtest

  • fold_info_df (DataFrame) – Fold information from backtest

log_backtest_run(metrics: DataFrame, forecast: DataFrame, test: DataFrame)[source]#

Backtest metrics from one fold to logger.

Parameters:

  • metrics (DataFrame) – Dataframe with metrics from backtest fold

  • forecast (DataFrame) – Dataframe with forecast

  • test (DataFrame) – Dataframe with ground truth

reinit_experiment()[source]#

Reinit experiment.

set_params(**params: dict) → Self[source]#

Return new object instance with modified parameters.

Method also allows to change parameters of nested objects within the current object. For example, it is possible to change parameters of a model in a Pipeline.

Nested parameters are expected to be in a <component_1>.<...>.<parameter> form, where components are separated by a dot.

Parameters:

**params (dict) – Estimator parameters

Returns:

New instance with changed parameters

Return type:

Self

Examples

>>> from etna.pipeline import Pipeline
>>> from etna.models import NaiveModel
>>> from etna.transforms import AddConstTransform
>>> model = NaiveModel(lag=1)
>>> transforms = [AddConstTransform(in_column="target", value=1)]
>>> pipeline = Pipeline(model, transforms=transforms, horizon=3)
>>> pipeline.set_params(**{"model.lag": 3, "transforms.0.value": 2})
Pipeline(model = NaiveModel(lag = 3, ), transforms = [AddConstTransform(in_column = 'target', value = 2, inplace = True, out_column = None, )], horizon = 3, )
start_experiment(job_type: str | None = None, group: str | None = None, *args, **kwargs)[source]#

Start experiment.

Complete logger initialization or reinitialize it before the next experiment with the same name.

Parameters:

  • job_type (str | None) – Specify the type of run, which is useful when you’re grouping runs together into larger experiments using group.

  • group (str | None) – Specify a group to organize individual runs into a larger experiment.

to_dict()[source]#

Collect all information about etna object in dict.

property experiment[source]#

Init experiment.

property pl_logger[source]#

Pytorch lightning loggers.