Closed helloobaby closed 11 months ago
hi @helloobaby , I can understand that this part can be confusing. Those are just some magic numbers that I approximated basing on my tests, it's not something precise. Just ratios of characters that are common for particular type of content. Now it is hardcoded, but it may be replaced with some machine learning in the future.
Thank you for the explanation
I feel that your project is very useful, and I am also puzzled by some of your algorithms.
https://github.com/hasherezade/pe-sieve/blob/be19198456e562143e44ecc34e7d034184f4d42f/stats/stats_analyzer.cpp#L179
Do you have more detailed information?