Tom Ramcigam (magicmarmot) wrote,
Tom Ramcigam
magicmarmot

I am simply exhausted today. Not enough sleep last night, I hate hotel beds. And tonight I'm pretty much packing stuff up to take out to the truck.

Lots of work today. Kind of the norm-- very full days, and it's only going to get more busy as we go.
Helmut (the "new" guy) is getting a good handle on the system. We've been batting some ideas and questions back and forth, which is good because it's so new to both of us.

I have a big design-type question that has to do with the mention of a "best-fit line", so this is probably aimed at revdj more than anybody:

I have a series of points in a plane.
One point is defined as the center of a circle.
The rest of the points are contained within the circle.
The number of points is not known, but it is finite.
using the center point as the endpoint of a line segment, I need to define a single line that is a "best fit" for the remaining points.
However, the points are weighted. The points farther out toward the edge of the circle are "heavier". The weighting is somewhat arbitrary, but for the purpose of this analogue, the weighting can be considered linear with radial distance from the center point.

What I end up needing is the direction of the "best fit" line that passes through the center point.

My current plan is to create a vector from the center to each point, then extract a "mean" vector from all of those.

Is there an easier way?
Subscribe
  • Post a new comment

    Error

    default userpic

    Your reply will be screened

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.
  • 3 comments