-
Notifications
You must be signed in to change notification settings - Fork 18
Comparing changes
Open a pull request
base repository: PecanProject/sipnet
base: master
head repository: PecanProject/sipnet
compare: DoubleExponential
- 6 commits
- 72 files changed
- 1 contributor
Commits on Jun 5, 2019
-
Exploring a change the cost function in paramchange.c
I'm hoping to change the function from one assuming a gaussian distribution to one assuming a double exponential distribution of errors. git-svn-id: svn://cgdm-bishorn.wlclient.ucar.edu/Users/sacks/temporary/sipnet.svn/branches/DoubleExponential@67 c88c3319-8f09-0410-a662-fffcf0821eeb
Configuration menu - View commit details
-
Copy full SHA for 8bb6b58 - Browse repository at this point
Copy the full SHA 8bb6b58View commit details -
SIPNET Dave Moore's copy 08162007
git-svn-id: svn://cgdm-bishorn.wlclient.ucar.edu/Users/sacks/temporary/sipnet.svn/branches/DoubleExponential@68 c88c3319-8f09-0410-a662-fffcf0821eeb
Configuration menu - View commit details
-
Copy full SHA for 008659f - Browse repository at this point
Copy the full SHA 008659fView commit details -
absDifference[dataNum] = 0.0; //DM 08/16/2007 sumData[dataNum] = 0.0;//DM 08/19/2007 absSqDevMean = 0.0;//DM 08/19/2007 absDifference = the absolute difference between model and the data sumData = the sum of the data type absSqDevMean = the squared difference between each data measure and the mean of that data type Added in an estimate of Beta from Hollinger and Richardson 2005 Question: I'm not sure that the sigma calculation in the paramchange.c code is correct - sigma is a measure of the variability of the data. It should not be influenced by any modeled value The cost function as written contains deviations between the model and the data in both the numerator and denominator I'm not sure what effect this has on the optmization. git-svn-id: svn://cgdm-bishorn.wlclient.ucar.edu/Users/sacks/temporary/sipnet.svn/branches/DoubleExponential@69 c88c3319-8f09-0410-a662-fffcf0821eebConfiguration menu - View commit details
-
Copy full SHA for e4f281e - Browse repository at this point
Copy the full SHA e4f281eView commit details -
Including H&R 2005 double exponential in the cost function
git-svn-id: svn://cgdm-bishorn.wlclient.ucar.edu/Users/sacks/temporary/sipnet.svn/branches/DoubleExponential@70 c88c3319-8f09-0410-a662-fffcf0821eeb
Configuration menu - View commit details
-
Copy full SHA for 999788f - Browse repository at this point
Copy the full SHA 999788fView commit details -
cleared up some compiling errors
Question - sigma is not declared in paramchange.c ... I'm not sure where it is declared. But I needed to declare beta in paramchange.c git-svn-id: svn://cgdm-bishorn.wlclient.ucar.edu/Users/sacks/temporary/sipnet.svn/branches/DoubleExponential@71 c88c3319-8f09-0410-a662-fffcf0821eeb
Configuration menu - View commit details
-
Copy full SHA for 8488188 - Browse repository at this point
Copy the full SHA 8488188View commit details -
Previous attempt to reformulate the cost function led to a segmentati…
…on fault This attempt is simpler sumSquares has been replaced by absDifference in both the numerator and the denominator of the cost function git-svn-id: svn://cgdm-bishorn.wlclient.ucar.edu/Users/sacks/temporary/sipnet.svn/branches/DoubleExponential@72 c88c3319-8f09-0410-a662-fffcf0821eeb
Configuration menu - View commit details
-
Copy full SHA for 69ddd5d - Browse repository at this point
Copy the full SHA 69ddd5dView commit details
This comparison is taking too long to generate.
Unfortunately it looks like we can’t render this comparison for you right now. It might be too big, or there might be something weird with your repository.
You can try running this command locally to see the comparison on your machine:
git diff master...DoubleExponential