Rank: 1063 based on 88 downloads (last 30 days) and 10 files submitted
photo

Stephan Koehler

E-mail
Company/University
WPI
Lat/Long
42.26300048828125, -71.802001953125

Personal Profile:
Professional Interests:

 

Watch this Author's files

 

Files Posted by Stephan View all
  • 1 - 5 of
  • 10
Updated   File Tags Downloads
(last 30 days) 		Comments Rating
19 Apr 2013 Screenshot binning a point cloud, 3D scattered data, in the X-Y plane Take a set of XYZ points, and returns average Z values for corresponding bins in X & Y planes. Author: Stephan Koehler image processing, signal processing, statistics 31 1
21 Mar 2013 modified cumsum for partial accumulated sums Determine the accumulated sum of series of non-zero values of vectors, resetting at zero entries Author: Stephan Koehler mathematics, physics, matrix 8 0
07 Feb 2013 Screenshot Error Propagation Numerically calculates uncertainties of a function using random numbers to simulate function inputs Author: Stephan Koehler error analysis, error propagation 13 0
  • 5.0
5.0 | 1 rating
09 Feb 2012 fitting sum of two decaying exponentials (over damped oscillator) five-parameter fit to the equation A*exp(-t/T1) + B*exp( -t/T2 ) + offset Author: Stephan Koehler fit, overdamped oscillator 5 0
09 Feb 2012 fitting critically damped simple harmonic oscillator Four-parameter fit to the equation A*(1 + t/T2).*exp(-t/T1) + offset Author: Stephan Koehler fitting, harmonic oscillator 2 0
Comments and Ratings by Stephan View all
  • 1 - 5 of
  • 5
Updated File Comments Rating
05 May 2012 Generalized Nonlinear Non-analytic Chi-Square Fitting Performs chi-square fit with uncertainty estimation when measurement errors are known. Author: Nathaniel Brahms

Question: Could you also provide 90% confidence intervals for the fitting parameters, instead of just the 68%?
05 May 2012 Generalized Nonlinear Non-analytic Chi-Square Fitting Performs chi-square fit with uncertainty estimation when measurement errors are known. Author: Nathaniel Brahms

20 Aug 2011 Linear Regression with Errors in X and Y Calculates slope and intercept for linear regression of data with errors in X and Y. Author: Travis Wiens

02 Mar 2011 Iterative Adaptive Simpson and Lobatto Quadrature Iterative Adaptive Simpson and Lobatto Quadrature Author: Matthias Chung

this works much faster than MATLAB's quadv, because quadv does not vectorize (i.e. it evaluates the function using only single values).

05 Nov 2010 RMSEARCH Randomly multiply started optimizations for global problems Author: John D'Errico

works great, but there is one problem. I get an error
"Maximum number of function evaluations has been exceeded..."
How do you set the option for function evaluations here?
thanks!
Comments and Ratings on Stephan's Files View all
  • 1 - 5 of
  • 5
Updated File Comment by Comments Rating
20 Apr 2013 binning a point cloud, 3D scattered data, in the X-Y plane Take a set of XYZ points, and returns average Z values for corresponding bins in X & Y planes. Author: Stephan Koehler yagnesh

is it usefull for 3D face recognition?
09 Oct 2012 Error Propagation Numerically calculates uncertainties of a function using random numbers to simulate function inputs Author: Stephan Koehler Kresten

24 Nov 2009 new_dir.m This is an enhancement to MATLAB's dir function, for dealing with sequentially numbered files. Author: Stephan Koehler Simon, Jan

Bugs are fixed, descriptive warning on Windows for the -find flag, updated description on the FEX page (removed "Matlab 5.3"), considering directories, some improvements in the help section. No comments in the source. Useful and usable.
18 Nov 2009 new_dir.m This is an enhancement to MATLAB's dir function, for dealing with sequentially numbered files. Author: Stephan Koehler Simon, Jan

The function starts with overwriting VARARGIN - please remove this!
The help section now mentions, that the -find option works on Unix only. But a check in the function (if ~isunix, error('-find works on Unix only'); end) would be useful.
Please update the file information on the FEX: This function does still not run under Matlab 5.3.
The help section is still hard to read (for me) due to the formatting. A H1-line would be helpful (see LOOKFOR command).
Some simplifications are possible, e.g.: "numel({d.name})" equals "numel(d)",
"to_do = strncmp( varargin, '-', 1 );
neg_queery = ~to_do&strncmp(varargin, '~', 1 );" is easier as: "neg_queery = strncmp(varargin, '~', 1);".
Comments in the source are recommended generally.
I really like the idea of expanding DIR with a filter for numerical indices.

08 Nov 2009 new_dir.m This is an enhancement to MATLAB's dir function, for dealing with sequentially numbered files. Author: Stephan Koehler Simon, Jan

In contrary to the file informations above, new_dir does not run on Matlab 5.3.1 (and 6.5 also), because the DIR command does not reply the field 'datenum' for this versions.
No meaningful H1 line, the help text is long and exhaustive, but hard to read (please insert a space after the %, limit the line width, use some empty lines to divide sections). A lot of examples are given, but some crash: There is no "-regexp" flag for the FIND command of Windows XP --- perhaps your function is written for Unix only?! If so, please mention this.
No comments in the source. The initial section for debugging is useless for FEX users.
If updates of this files are submitted or searched, it is helpful to mention the full name of the author and the exact date of submission.
Working partially on WindowsXP, but not under the claimed Matlab version.
Top Tags Applied by Stephan
  • 1 - 5 of
  • 21
data import, fitting, signal processing, csvread, data management
Files Tagged by Stephan View all
  • 1 - 5 of
  • 10
Updated   File Tags Downloads
(last 30 days) 		Comments Rating
19 Apr 2013 Screenshot binning a point cloud, 3D scattered data, in the X-Y plane Take a set of XYZ points, and returns average Z values for corresponding bins in X & Y planes. Author: Stephan Koehler image processing, signal processing, statistics 31 1
21 Mar 2013 modified cumsum for partial accumulated sums Determine the accumulated sum of series of non-zero values of vectors, resetting at zero entries Author: Stephan Koehler mathematics, physics, matrix 8 0
07 Feb 2013 Screenshot Error Propagation Numerically calculates uncertainties of a function using random numbers to simulate function inputs Author: Stephan Koehler error analysis, error propagation 13 0
  • 5.0
5.0 | 1 rating
09 Feb 2012 fitting sum of two decaying exponentials (over damped oscillator) five-parameter fit to the equation A*exp(-t/T1) + B*exp( -t/T2 ) + offset Author: Stephan Koehler fit, overdamped oscillator 5 0
09 Feb 2012 fitting critically damped simple harmonic oscillator Four-parameter fit to the equation A*(1 + t/T2).*exp(-t/T1) + offset Author: Stephan Koehler fitting, harmonic oscillator 2 0

Contact us