I'm working on using Tensorflow to improve Chinese OCR. So far I collected 1500 fonts. Now I'm exporting all the glyphs, and I ran out of disk space yesterday.

I'm working on using Tensorflow to improve Chinese OCR. So far I collected 1500 fonts. Now I'm exporting all the glyphs, and I ran out of disk space yesterday. Then I'll upload the data set of all 75,000 characters here on HN where people can play with it. Before you say that OCR is a solved problem because of Tesseract, please read this: https://ocr.space/blog/2015/03/best-ocr-software-for-chinese.html https://ocr.space/blog/2015/03/best-ocr-software-for-chinese...