Needing ML Model Training Data
Hi everyone,
Currently working on a development project and have hit a bit of a wall. I’m trying to build a ML model optimized to parse text from printed Canadian receipts, but finding a localized dataset online has been impossible.
I’m now crowdsourcing some training data and would love your help. If you have any paper receipts lying around and wouldn’t mind snapping a quick, clear photo of them, and sending it to me, it would be a massive help.
The receipts should have merchant name, list of items and their individual costs, totals, subtotals, taxes.
Please feel free to blackout any sensitive info. You can drop it in the comments or send them directly to my inbox via DM.
For anyone in the software/data science space, I’d also love to hear any tips you might have on overcoming localized OCR hurdles.
Thanks everyone!!