Co-Founder at SiPhox (YC S20) (We're Hiring) Sauna Subject Matter Expert + Brazilian Jiu Jitsu White Belt
Catching up with friends and investors in SF = solving the traveling salesman (founder) problem [1]. Does anyone have a quantum computer or slime mold colony [2] I can borrow to find a hotel that optimizes travel time to Palo Alto, SF, and Berkeley weighted by timing and # of meetings? 1. https://lnkd.in/eetRz2Va 2. https://lnkd.in/e2TQy8cH
Ethan Garber
2mo
Hotwire
VP - Aptitude Medical Systems | Democratizing Dx
2moOh I love graph theory! I actually scripted almost this exact thing for JPM last year. Unsurprisingly it just told me to stay in Union Square, but we stayed in the Tenderloin because it was 10X cheaper (graph theory can't solve everything). So a waste of time, but fun! Also, what you're describing isn't actually a Traveling Salesman Problem because you can't change the order of the meetings, you're just trying to minimize path lengths between a pre set list. That's actually a lot less computationally intensive (and a poor choice for slime mold based computing). If you email me your meetings in this format (you can get lat/long from google maps) I'll run it tonight after I do my actual job... meetings = pd.DataFrame({ 'location': ['Location A', 'Location B', 'Location C', 'Location D', 'Location E', 'Location F', 'Location G'], 'latitude': [lat1, lat2, lat3, lat4, lat5, lat6, lat7], 'longitude': [lon1, lon2, lon3, lon4, lon5, lon6, lon7], 'time': [time1, time2, time3, time4, time5, time6, time7], 'day': [day1, day1, day1, day1, day2, day2, day2] })