Skip to content

fix(dataset): update batch_size in sampling_parameters_json during project top-ups - #1273

Merged
Dhara124 merged 1 commit into
back-branch-masterfrom
feature/create-dataset-and-project
Sep 24, 2026
Merged

Dhara124 merged 1 commit into
back-branch-masterfrom
feature/create-dataset-and-project

Conversation

@Dhara124

Copy link
Copy Markdown
Collaborator

Description

This PR fixes a bug in the automated dataset pipeline where project top-ups would result in stale metadata being displayed in the project settings UI.

Bug Fixes

  • Dynamic Batch Size Updates: Modified the create_projects_for_dataset_category logic inside dataset/tasks.py. Previously, when a pipeline dynamically topped up an existing project by pulling in new tasks, the sampling_parameters_json.batch_size field remained as a static snapshot taken during the original project creation. This caused the "Batch Size" field in the UI settings to go out-of-sync with the actual task count. The batch size is now accurately updated and kept in sync by adding the length of newly pulled items to the existing value.

Testing Instructions

  1. Upload a dataset and initiate the project creation pipeline. Note the initial batch size in the project's settings.
  2. Upload a new part (e.g. Part B) for the same domain to trigger an automatic project top-up.
  3. Check the project settings again. Verify that the "Batch Size" field correctly reflects the total number of tasks (original batch + topped-up batch) instead of just the original batch.

@Dhara124
Dhara124 merged commit 113373a into back-branch-master Sep 24, 2026
2 of 3 checks passed
@Dhara124
Dhara124 deleted the feature/create-dataset-and-project branch September 24, 2026 03:56
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant