Contouring 8 points
in [Matlab]
|
Prev: blood simscape
Next: neural network outputs
From: Bruno Luong on 26 Apr 2010 08:02 "Nadine Cooper" <f9071770(a)bournemouth.ac.uk> wrote in message <hr2gr2$e4v$1(a)fred.mathworks.com>... > "Bruno Luong" <b.luong(a)fogale.findmycountry> wrote in message <hqqevc$3h0$1(a)fred.mathworks.com>... > > "Nadine Cooper" <f9071770(a)bournemouth.ac.uk> wrote in message <hqq7qv$s18$1(a)fred.mathworks.com>... > > > > > > Is that better???????? lol > > > > Ah quantum leap, but you could also in a most natural way: > > > > z2 = z2*(100/0.0720 * 0.00750061505043) > > > > > > > > I hate programming!! :) > > > > You'll like it. > > > > Bruno > > I hang my head in shame! > > I'm going to bother you one last time Bruno, just because I'm curious: > > probability density function is f(x)= 1/(σ√(2π )) e(- ((x- μ)^2)/(2σ^2 )) > > which is what you're doing above - > > However, how does it work? I mean, so I've worked out that σ is 0.016, but why did you not include the sqrt2pi √(2π )? > > So a gaussian fit is done initially to set the peaks up, their "height" is how I think of it, with σ as 0.016, correlating to x values and adding the y values Yes, we could include the constant factor but this constant factor (1) can be arbitrary chosen if we change the (Borel)'s measure of the location space, (2) does not really matter because it is include in the inversion in order to provide the surface that interpolate the data. So why bother with the unimportant details? Bruno
From: Nadine Cooper on 26 Apr 2010 08:52
"Bruno Luong" <b.luong(a)fogale.findmycountry> wrote in message <hr3vbt$r49$1(a)fred.mathworks.com>... > "Nadine Cooper" <f9071770(a)bournemouth.ac.uk> wrote in message <hr2gr2$e4v$1(a)fred.mathworks.com>... > > "Bruno Luong" <b.luong(a)fogale.findmycountry> wrote in message <hqqevc$3h0$1(a)fred.mathworks.com>... > > > "Nadine Cooper" <f9071770(a)bournemouth.ac.uk> wrote in message <hqq7qv$s18$1(a)fred.mathworks.com>... > > > > > > > > Is that better???????? lol > > > > > > Ah quantum leap, but you could also in a most natural way: > > > > > > z2 = z2*(100/0.0720 * 0.00750061505043) > > > > > > > > > > > I hate programming!! :) > > > > > > You'll like it. > > > > > > Bruno > > > > I hang my head in shame! > > > > I'm going to bother you one last time Bruno, just because I'm curious: > > > > probability density function is f(x)= 1/(σ√(2π )) e(- ((x- μ)^2)/(2σ^2 )) > > > > which is what you're doing above - > > > > However, how does it work? I mean, so I've worked out that σ is 0.016, but why did you not include the sqrt2pi √(2π )? > > > > So a gaussian fit is done initially to set the peaks up, their "height" is how I think of it, with σ as 0.016, correlating to x values and adding the y values > > Yes, we could include the constant factor but this constant factor > (1) can be arbitrary chosen if we change the (Borel)'s measure of the location space, > (2) does not really matter because it is include in the inversion in order to provide the surface that interpolate the data. > > So why bother with the unimportant details? > > Bruno hmmm.....fair enough. I just thought I'd ask. I was just really impressed that you knew practically immediately what I needed to do (ie, Gaussian fit) to get the peak elevations. Because I didn't even KNOW about Gaussian fits before! So therefore, I think.... okay, I want to know more! But thanks Bruno, for all your help. You're a legend and then some! :) |