Datasets:

danish-foundation-models
/

swedish-dynaword

There are collection of Swedish OCR datasets, these datasets are in "line" format (so I don't know it is suitable for pre-training or not) - Riksarkivet/training-data-for-swedish-lion-libre - total around 6M tokens

Largest dataset from the collection is 350,000 rows and around 3.5M tokens- Riksarkivet/goteborgs_poliskammare_fore_1900_lines

There are also three datasets in "paragraph" - format, V4ldeLund/ocr-eval-datasets around 1M tokens together

mitanuriel

9 days ago

⚠️ Test results

Failed to set up the test environment:

fatal: couldn't find remote ref refs/pr/1/head

mitanuriel

8 days ago

⚠️ Test results

Failed to set up the test environment:

fatal: couldn't find remote ref refs/pr/1/head

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment