Blind Assist

Exporting Datasets

Exporting Datasets

Export your media and captions in formats compatible with popular AI training frameworks.

Export Formats

JSONL

One JSON object per line. Each object contains the image path, caption, and metadata.

{"image": "img_001.jpg", "caption": "A photo of..."}

CSV

Comma-separated values with headers. Compatible with spreadsheet applications.

Text-per-Image

One .txt file per image, containing only the caption. Standard format for many training scripts.

Filtering Before Export

Use the library filter to narrow down what gets exported:

  • By status -- only export approved items
  • By tag -- export specific subsets
  • By quality -- filter by blur score

Export Options

  • Include or exclude metadata
  • Choose caption field (primary, secondary, etc.)
  • Set filename patterns
An unhandled error has occurred. Reload X