OrgLion-ML
/
dlib
mirror of https://github.com/davisking/dlib.git
1
0
Fork 0
Code Issues Releases Wiki Activity

Made comments more clear. 

--HG--
extra : convert_revision : svn%3Afdd8eb12-d10e-0410-9acb-85c331704f74/trunk%403567
This commit is contained in:
Davis King 2010-04-23 20:52:18 +00:00
parent d02aaa1bde
commit 8ebae5b954
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
examples/optimization_ex.cpp
View File

@ -77,16 +77,16 @@ class test_function
In this example, our test_function object contains a column_vector In this example, our test_function object contains a column_vector
as its state and it computes the mean squared error between this as its state and it computes the mean squared error between this
stored column_vector and the arguments to its operator() function. stored column_vector and the arguments to its operator() function.
This is a very simple function. However, in general you could compute
This is a very simple function, however, in general you could compute
any function you wanted here. An example of a typical use would be any function you wanted here. An example of a typical use would be
to find the parameters to some regression function that minimized to find the parameters of some regression function that minimized
the mean squared error on a set of data. In this case the arguments the mean squared error on a set of data. In this case the arguments
to the operator() function would be the parameters of your regression to the operator() function would be the parameters of your regression
function and you would use those parameters to loop over all your data function. You would loop over all your data samples and compute the output
samples, compute the output of the regression function given those of the regression function for each data sample given the parameters and
parameters, and finally return a measure of the error. The dlib return a measure of the total error. The dlib optimization functions
optimization functions would then be used to find the parameters that could then be used to find the parameters that minimized the error.
minimized the error.
*/ */
public: public: