search

atlanhq / camelot

Camelot: PDF Table Extraction for Humans
https://camelot-py.readthedocs.io
Other
3.66k stars 360 forks source link

Negative accuracy #223

Closed anakin87 closed 5 years ago

anakin87 commented 5 years ago

FI Dimensione Investimento Aziende Mod. 220233 Ed. 05-2018.pdf

I extracted the tables from this file with this syntax: tables=camelot.read_pdf(filename, pages='5',flavor='lattice',line_size_scaling=65,copy_text=['h','v'])

tables[3].parsing_report shows me the following stats: {'accuracy': -2382.15, 'order': 4, 'page': 5, 'whitespace': 0.0}

The accuracy is negative. Can you reproduce this problem?

BTW, I would like to thank you for the wonderful work you are doing...

vinayak-mehta commented 5 years ago

Thanks for the report, I'll look into this.