-------- Original Message --------
Subject: Re: Missing Landmarks
Date: Wed, 1 Oct 2008 05:05:05 -0700 (PDT)
From: Dennis E. Slice <[EMAIL PROTECTED]>
To: [email protected]
References: <[EMAIL PROTECTED]>
Okay, okay. The new Morpheus et al. has some functioning features that I
have found very useful in cleaning data with missing values. Well, I
wrote them to check a 1000+ 3D data set with numerous scattered missing
values, and I use it all the time.
So, while I do not feel the program is ready for release, I don't want
to deny anyone who might find these functions useful. Therefore, I have
uploaded an undocumented, unsupported, pre-release alpha version. It can
be found here:
http://www.morphometrics.org/morpheus.html
Save the .jar file, from a command window change to that directory, run
the program with "java -jar morpheus_alpha_01_oct_2008.jar"
In general, you're on your own.
Some may find the following choices under the main menu LIST useful:
POINTREPORT
MISSINGREPORT
Output (and other functionality) should be self-explanatory.
While this version is unsupported, I do encourage and welcome comments,
contributions, and bug reports that I will put into my "suggestion box"
for consideration when I get back to coding.
Best, ds
PS: If you wish, you can imagine the m_vis graphics window in the
current crummy graphics window in the alpha version. That will give you
an idea of the final product.
morphmet wrote:
-------- Original Message --------
Subject: RE: Missing Landmarks
Date: Tue, 30 Sep 2008 05:10:40 -0700 (PDT)
From: Steven Wang <[EMAIL PROTECTED]>
To: <[email protected]>
References: <[EMAIL PROTECTED]>
To follow up on Michelle's comment, Morpheus recognizes 9999 as missing
data, so my template always starts off with all 9999 (for x, y, and z).
I would then simply digitize actual data over each column in Excel. 9999
is also an "outrageous coordinate," so it is very easy to catch in your
dataset.
That's during digitization... but, trying to locate missing data during
analysis for every single specimen is a whole different level of PAIN. I
use SAS, but it's quite complicated (requires knowledge of SAS
programming). So I am interested in what others have to say on this issue.
Cheers,
Steve
CUNY/NYCEP
--
Dennis E. Slice
Dept. of Scientific Computing
Florida State University
Dirac Science Library
Tallahassee, FL 32306-4120
-
Department of Anthropology
University of Vienna
========================================================
--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org
