dacapo.experiments.datasplits.datasplit_generator

Module Contents

Classes

CustomEnumMeta	Metaclass for Enum
CustomEnum	Generic enumeration.
DatasetType	Generic enumeration.
SegmentationType	Generic enumeration.
DatasetSpec
DataSplitGenerator	Generates DataSplitConfig for a given task config and datasets.

Functions

is_zarr_group(file_name, dataset)
resize_if_needed(array_config, target_resolution[, ...])
get_right_resolution_array_config(container, dataset, ...)
generate_dataspec_from_csv(csv_path)
format_class_name(class_name[, separator_character])

Attributes

logger

dacapo.experiments.datasplits.datasplit_generator.logger

dacapo.experiments.datasplits.datasplit_generator.is_zarr_group(file_name: str, dataset: str)

dacapo.experiments.datasplits.datasplit_generator.resize_if_needed(array_config: dacapo.experiments.datasplits.datasets.arrays.ZarrArrayConfig, target_resolution: funlib.geometry.Coordinate, extra_str='')

dacapo.experiments.datasplits.datasplit_generator.get_right_resolution_array_config(container: pathlib.Path, dataset, target_resolution, extra_str='')

class dacapo.experiments.datasplits.datasplit_generator.CustomEnumMeta

Metaclass for Enum

class dacapo.experiments.datasplits.datasplit_generator.CustomEnum

Generic enumeration.

Derive from this class to define new enumerations.

class dacapo.experiments.datasplits.datasplit_generator.DatasetType

Generic enumeration.

Derive from this class to define new enumerations.

val = 1

train = 2

class dacapo.experiments.datasplits.datasplit_generator.SegmentationType

Generic enumeration.

Derive from this class to define new enumerations.

semantic = 1

instance = 2

class dacapo.experiments.datasplits.datasplit_generator.DatasetSpec(dataset_type: str | DatasetType, raw_container: str | pathlib.Path, raw_dataset: str, gt_container: str | pathlib.Path, gt_dataset: str)

dacapo.experiments.datasplits.datasplit_generator.generate_dataspec_from_csv(csv_path: pathlib.Path)

class dacapo.experiments.datasplits.datasplit_generator.DataSplitGenerator(name: str, datasets: List[DatasetSpec], input_resolution: funlib.geometry.Coordinate, output_resolution: funlib.geometry.Coordinate, targets: List[str] | None = None, segmentation_type: str | SegmentationType = 'semantic', max_gt_downsample=32, max_gt_upsample=4, max_raw_training_downsample=16, max_raw_training_upsample=2, max_raw_validation_downsample=8, max_raw_validation_upsample=2, min_training_volume_size=8000, raw_min=0, raw_max=255, classes_separator_caracter='&')

Generates DataSplitConfig for a given task config and datasets. class names in gt_dataset shoulb be within [] e.g. [mito&peroxisome&er] for mutiple classes or [mito] for one class Currently only supports:

• semantic segmentation.

Supports:

• 2D and 3D datasets.

• Zarr, N5 and OME-Zarr datasets.

• Multi class targets.

property class_name

check_class_name(class_name)

compute()

static generate_from_csv(csv_path: pathlib.Path, input_resolution: funlib.geometry.Coordinate, output_resolution: funlib.geometry.Coordinate, name: str | None = None, **kwargs)

dacapo.experiments.datasplits.datasplit_generator.format_class_name(class_name, separator_character='&')