Found this from https://assets.ctfassets.net/we0wnls2rnwy/2xu8qs3xCoWjFLe70tmzse/00b29fef3b7ee0abd61d9fd84d4a1786/Evolv_HS.pdf https://assets.ctfassets.net/we0w

Found this from https://assets.ctfassets.net/we0wnls2rnwy/2xu8qs3xCoWjFLe70tmzse/00b29fef3b7ee0abd61d9fd84d4a1786/Evolv_HS.pdf https://assets.ctfassets.net/we0wnls2rnwy/2xu8qs3xCoWjFLe70t... which I found on https://www.jsx.com/faqs https://www.jsx.com/faqs what 100% better way to do it! "Using large databases of threats, including firearms, metallic IEDs, and other threats as well as databases of typical benign objects, including keys, cellphones, belts, shoes, coins, tablets, and laptops, the classifier is trained to distinguish between the two classes via machine learning techniques"