GutenTAG has a simple add-on feature which can be activated by using the CLI.
An add-on can be written and defined with the help of the BaseAddOn
class.
This class defines two methods, that are called after the CLI tool has already saved the time series to disk:
import os
from typing import List, Optional
from dataclasses import dataclass
from gutenTAG import GutenTAG
from gutenTAG.generator import Overview, TimeSeries
@dataclass
class AddOnProcessContext:
overview: Overview
datasets: List[TimeSeries]
plot: bool = False
output_folder: Optional[os.PathLike] = None
n_jobs: int = 1
class BaseAddOn:
def process(self, ctx: AddOnProcessContext, gutenTAG: GutenTAG) -> AddOnProcessContext:
"""Gets called after time series are generated but before they are plotted or written to disk."""
return ctx
GutenTAG already comes with an add-on for the TimeEval tool.
This add-on creates a datasets.csv
meta-file in the output directory.
This file is necessary to import the generated datasets into the TimeEval tool.
Use it as follows:
python -m gutenTAG --config-yaml config.yaml --addons gutenTAG.addons.timeeval.TimeEvalAddOn