Datasets:
Tasks:
Image-to-Text
Modalities:
Text
Formats:
webdataset
Languages:
English
Size:
1K - 10K
License:
Non-English language pdfs
#5
by
NoOneCodesAI
- opened
Hello,
Thanks for the great dataset. It is extremely handy. Is there any possibility to obtain a version that has NOT been filtered by language? I would like to asses OCR performance across different languages as this seems to make quite a difference.
Thanks in advance!
Thanks for your reply! I would love to hear about the status, the unfiltered dataset would be perfect for a project I want to work on :)