Plexora
Python APIDatasets

Dataset

A named group of projects, and a handle for changing that group.

Generated from Plexora 0.0.55View source
from plexora import Dataset
Dataset(
    id: str,
    name: str,
    description: str = '',
    created_at: str = '',
    projects: tuple[str, ...] = (),
    meta: Mapping[str, Any] = dict(),
)

Returned by create_dataset, dataset and list_datasets -- never constructed directly. Supports len(ds), for name in ds and "slide1" in ds over its project names.

Every method below returns a fresh Dataset rather than changing this one in place: another process (the viewer, another notebook) can write to the same dataset at any time, so a handle is a snapshot from the moment it was read. Reassign the result to keep working with the current state, e.g. cohort = cohort.add("slide3").

Fields

idstrrequired

The dataset's id. Stable even if the dataset is renamed.

namestrrequired

The dataset's name, as shown in the UI and passed to dataset.

descriptionstrdefault ''

A free-text note about the dataset. Empty if none was set.

created_atstrdefault ''

When the dataset was created, as an ISO timestamp.

projectstuple[str, ...]default ()

The names of the projects currently in this dataset.

metaMapping[str, Any]default dict()

Other metadata set with describe(**meta).

Methods

__len__

__len__()

The number of projects in this dataset.

__iter__

__iter__()

Iterate over the dataset's project names.

__contains__

__contains__(name)

Whether name is one of this dataset's projects.

namestrrequired

The project name to look for.

refresh

refresh() -> 'Dataset'

Reload this dataset and return the current handle.

Picks up anything another process changed since this handle was made -- a project added from the viewer, say, while a notebook still holds the dataset from before that.

Dataset

The dataset as it is now.

add

add(*names) -> 'Dataset'

Put projects into this dataset, taking them out of any other one.

*names

Project names, or a mix of names and lists of names.

Dataset

The updated dataset.

remove

remove(*names) -> 'Dataset'

Take projects out of this dataset. They are not deleted.

*names

Project names, or a mix of names and lists of names. Names not currently in this dataset are ignored.

Dataset

The updated dataset.

rename

rename(name) -> 'Dataset'

Rename this dataset.

namestrrequired

The new name.

Dataset

The renamed dataset.

describe

describe(text=None, **meta) -> 'Dataset'

Set the dataset's description, its metadata, or both.

textstrdefault None

The new description. Leave unset to keep the current one.

**meta

Metadata to merge into the dataset's existing meta.

Dataset

The updated dataset.

delete

delete() -> None

Delete this dataset.

The projects in it are not deleted -- they stay registered at the top level, exactly as deleting a folder in the UI leaves its files behind.

manifest

manifest() -> dict

What every project in this dataset has, and what is still open.

The same per-project readiness check project_manifest makes, run once for each project in the dataset.

dict

{project_name: manifest}, one entry per project, in the shape project_manifest returns.

Examples

import plexora

cohort = plexora.create_dataset("Melanoma Cohort", images=["slide.ome.tif"])
print(cohort.name, list(cohort))

How is this different?

dataset or Dataset?

plexora.dataset(name) is the function that looks a dataset up. plexora.Dataset is the class of what it returns. You rarely construct a Dataset yourself.

create_dataset

Make a dataset, registering its projects if they are not there yet.

dataset

Get one dataset, by name or by id.

Guides

Opening images and datasets

dataset

Get one dataset, by name or by id.

DatasetCreateError

Raised by create_dataset when a batch registration fails part-way.

On this page

FieldsMethods__len____iter____contains__refreshaddremoverenamedescribedeletemanifestExamplesHow is this different?RelatedGuides