Tim Wicinski @billjelavich.bsky.social

Textract - interesting! I will need to look into this. This means you can process 2024 data quicker i take it?

2 replies 0 reposts 1 likes

Derek Willis @dwillis.bsky.social [ View ] Jan 28, 2024 at 01:05 UTC

Yes, especially for those states where counties produce image PDFs.

0 replies 0 reposts 1 likes

Lisa Pickoff-White @pickoffwhite.bsky.social [ View ] Jan 28, 2024 at 16:33 UTC

I’ve also found Textract to be very efficient at handling narrative and form text on the same page at scale.

This is a handy guide to different OCR engines rn source.opennews.org/articles/our...

0 replies 1 reposts 2 likes