195 Reputation

7 Badges

11 years, 239 days

MaplePrimes Activity

These are replies submitted by AndrewG

@Markiyan Hirnyk I see this question was asked a lot but none of the answers were satisfying in my opinion.  They weren't elagantly simple. 

Isn't it strange Maple doesn't have a simple convert to scientific notation command?

@Kitonum Thanks, never thought about using Search.


Using cat is similar but doesn't need typeset

plot(sin((1/180)*Pi*x), x = -180 .. 180, tickmarks = [[seq(180*i*(1/4) = cat(180*i*(1/4), `°`), i = -4 .. 4)], default])

@Carl Love k means clustering tends to be less complex and faster and considered superior for some reason.  Fuzzy c means is weighted giving more importance to stronger connected points.

Actually I didn't realize there were so many different types of clustering algorithms or different names attached to them.  The fuzzy c-means clustering is also known as soft k means.

It would them seem redundant to use RandomVariable command in the Sample statement.

@vv thanks.  I see from the help page description

The Sample command generates a random sample drawn from the distribution given by X.

The RandomVariable command creates new random variable with the specified distribution.

@Carl Love Thanks for the work on that. 

I found a post on k mean clustering that was open ended.  It appears we should be able to extrapolate this code to k mean clustering?

@Markiyan Hirnyk Thanks for your example.  You have done quite a bit of work. 


@tomleslie thanks for the alternate.  I was hoping someone would show another way, I didn't know if the way I did it was a correct way or if maple had a command I didn't know about. 

Mathematica (which I don't know much about) creates it like this



To me it looks like they generate a normal distribution around the point 0,0 and then just add it to a new point.  Isn't this the same thing



So you're saying what I created is not a general normal distribution around a specific point?  Show me how you would do it. 


If X and Y are normally distributed and independent (they are), this implies they are "jointly normally distributed". 

@Markiyan Hirnyk I probably should have been more specific.  I appreciate you looking into it.

Regardless, thank you Markiyan for pointing me to I came across a help page that gave me an idea which produced my above result. 

So the way I created the points, are they still considered distributed normally about point [3,15] ?  or have I missed something?


@Markiyan Hirnyk just a two dimensional distribution

@Carl Love ok here is some sample data


1 2 3 4 5 Page 2 of 5