split_dataset#

gammapy.datasets.split_dataset(dataset, width, margin, split_template_models=True)[source]#

Split dataset in multiple non-overlapping analysis regions.

Parameters:

datasetDataset: Dataset to split.
widthAngle: Angular size of each sub-region.
marginAngle: Angular size to be added to the width. The margin should be defined such as sources outside the region of interest that contributes inside are well-defined. The mask_fit in the margin region is False and unchanged elsewhere.
split_template_modelsbool, optional: Apply cutout to template models or not. Default is True.

Returns:

datasetsDatasets: Split datasets.

Examples

>>> from gammapy.datasets import MapDataset
>>> from gammapy.datasets.utils import split_dataset
>>> from gammapy.modeling.models import (
...     GaussianSpatialModel,
...     PowerLawSpectralModel,
...     SkyModel,
... )
>>> import astropy.units as u
>>> dataset = MapDataset.read("$GAMMAPY_DATA/cta-1dc-gc/cta-1dc-gc.fits.gz")
>>> # Split the dataset
>>> width = 4 * u.deg
>>> margin = 1 * u.deg
>>> split_datasets = split_dataset(
...     dataset, width, margin, split_template_models=False
... )
>>> # Apply a model and split the dataset
>>> spatial_model = GaussianSpatialModel()
>>> spectral_model = PowerLawSpectralModel()
>>> sky_model = SkyModel(
...     spatial_model=spatial_model,
...     spectral_model=spectral_model,
...     name="test-model",
... )
>>> dataset.models = [sky_model]
>>> split_datasets = split_dataset(
...     dataset, width=width, margin=margin, split_template_models=True
... )