Skip to content
ModAsteraDocumentation
日本語
Datasets · Cross-dataset workspace

Work across datasets in All Samples

Find samples across the workspace, select a compatible scope, create or extend a dataset, and delete only confirmed samples.

Browse and select samples across datasets#

Before you begin

  • Clear any prior selection before starting a task with a different data type or destination.
  1. Open All Samples

    Review the visible sample previews, names, types, and source context.

  2. Narrow the workspace

    Use available search, filters, and pagination to reach the intended samples.

  3. Select compatible items

    Choose individual samples, or select all samples on the current page. Do not assume that page selection includes every result on other pages.

  4. Confirm one data type

    Before creating or extending a dataset, ensure all selected samples have the same compatible type.

  5. Review the selection tray

    Confirm the count and intended action before continuing.

All Samples workspace with cross-dataset results, sample selection, and bulk action controls
Select all applies to the current page; review type and count before a bulk action.

Create a dataset from selected samples#

Before you begin

  • Confirm the exact selected count and data type.
  • Prepare a non-sensitive dataset name and description.
  1. Choose Create Dataset

    Start the action from the selection tray after confirming the selected samples.

  2. Name and describe the destination

    Enter a clear name and description that explain why these samples form a dataset.

  3. Create once

    Submit the operation and wait for the destination to be created.

  4. Open and verify

    Open the new dataset and compare its type and sample count with the original selection.

Add selected samples to an existing dataset#

Before you begin

  • Confirm the selected samples share one type.
  • Know whether the destination splits are locked.
  1. Choose Add to Dataset

    Open the existing-dataset action from the selection tray.

  2. Find a compatible destination

    Search for and select the intended editable dataset. Clear the search if the expected destination is hidden.

  3. Set optional metadata

    Choose an available split and, when needed, a classification label and value to apply to the added samples.

  4. Respect a locked test split

    Do not assign new samples to Test when the destination split is locked; choose another split or leave it unset.

  5. Submit and reconcile

    Wait for completion, open the destination, and review added, duplicate, skipped, or failed samples.

Delete selected samples safely#

Before you begin

  • Confirm every selected sample and the current page or filter scope.
  • Preserve any source or annotations needed for recovery or audit.
  1. Review the selection

    Inspect the selection tray count and open uncertain samples before choosing Delete.

  2. Read the confirmation

    Verify that the dialog names the intended action and sample scope. Cancel when either is unclear.

  3. Confirm once

    Proceed only after the selected items and consequences are understood.

  4. Verify affected datasets

    Refresh All Samples and any source datasets, then recheck counts, splits, readiness, and dependent work.

Continue with

Create a dataset and add samplesChoose the right creation path, upload or register supported data, and follow asynchronous processing to completion.Find, inspect, and edit dataset samplesUse split tabs, search, filters, card, list, and table views, then inspect processing and edit only the intended sample scope.Manage splits, collaboration, and advanced dataset actionsCreate reproducible splits, protect test membership, clone safely, manage access, and use inference or DICOM extraction when eligible.