Is it configuration way to obtain The better filding?

[Alex-Jud]

I show two parameter used for test more solutions. EXTRA_SAMPLES && TRIES_PER_TRI
Can we add an option to test all options?

[jpeterbaker]

It's very slow to try every combination of links, and I haven't come up with a good way to iterate through them yet. I still want this in a future update.

[Alex-Jud]

I'm looking to try something for find the best segments arrangement, if it work fine I tell you.
thks

[mrplow]

I generally run a script to run the maxfield script 1000 times and then I
grep |sort|cut et all the results to find the plan with the shortest
walking distance

On Wed, Jul 9, 2014 at 3:43 PM, Alex-Jud [email protected] wrote:

I'm looking to try something for find the best segments arrangement, if it
work fine I tell you.
thks

—
Reply to this email directly or view it on GitHub
#15 (comment).

[Alex-Jud]

I am developing a Python script to generate the best field. I look to incorporate my result in your script.
In fact, my script can generate an nx.graph like yours; loading nodes name , keys, geo , xy , xyz from a csv file and add the edges having the best fielding (ok it takes two hours of calculation for 20 portals)
But my script does not take care of the order , number key, or anything else.
Is there a way to incorporate my external edges in maxfield so he can do the rest of work?

[rddaz2013]

• (ok it takes two hours of calculation for 20 portals)
• It's very slow to try every combination of links

more than 15+ Portal it is very hard to get a result, because of the method of 'recombination'.

perhaps we don't search for the best fields only a better field.