0
|
5911 Views
|
|
1 Total Likes
View groups...
Share
GROUPS:

Understand the NMinimize output vs. StepMonitor output

Posted 10 years ago
 Here i s part of my code. I am looking for a minimal value of a function krit (with a given parameter Fexp). The output from the StepMonitor says, that NMinimize found a value 87.985 of the function krit, but the result of the NMinimize function is (worse) value 99.8154. Why? Why not 87.985? Thank you for help! In[77]:= res = NMinimize[ krit[k1, k2, n1, n2, \[CapitalDelta]1, \[CapitalDelta]2, e1, e2, Fexp], k1 > 0 && k2 > 0 && n1 > 0.01 && n2 > 0.01 && \[CapitalDelta]1 > 0 && \[CapitalDelta]2 > 0 && e1 > 0.01 && e2 > 0.01 && \[CapitalDelta]1 \[Element] Integers && \[CapitalDelta]2 \[Element] Integers, {k1, k2, n1, n2, \[CapitalDelta]1, \[CapitalDelta]2, e1, e2}, MaxIterations -> 5, StepMonitor :> {Print[ krit[k1, k2, n1, n2, \[CapitalDelta]1, \[CapitalDelta]2, e1, e2, Fexp], " <-> ", {k1, k2, n1, n2, \[CapitalDelta]1, \[CapitalDelta]2, e1, e2}]} ] During evaluation of In[77]:= 99.9059 <-> {1.73071,1.13718,0.601198,1.31724,2,2,1.05854,0.129745} During evaluation of In[77]:= 99.9059 <-> {1.73071,1.13718,0.601198,1.31724,2,2,1.05854,0.129745} During evaluation of In[77]:= 99.9059 <-> {1.73071,1.13718,0.601198,1.31724,2,2,1.05854,0.129745} During evaluation of In[77]:= 99.8549 <-> {1.85288,1.36396,0.739199,1.1769,4,1,1.07687,0.807171} During evaluation of In[77]:= 99.8549 <-> {1.85288,1.36396,0.739199,1.1769,4,1,1.07687,0.807171} During evaluation of In[77]:= 99.8154 <-> {1.37364,1.49254,0.147263,1.26496,4,1,1.01063,1.75675} During evaluation of In[77]:= 96.4886 <-> {1.53655,1.49543,0.0669391,1.26766,4,1,0.299039,1.75928} During evaluation of In[77]:= 88.3542 <-> {1.58294,1.49545,0.026076,1.26766,4,1,0.0128904,1.75928} During evaluation of In[77]:= 88.1938 <-> {1.58489,1.49545,0.0242414,1.26766,4,1,0.0100289,1.75928} During evaluation of In[77]:= 87.985 <-> {1.60044,1.49546,0.0109273,1.26766,4,1,0.0100003,1.75929} Out[77]= {99.8154, {k1 -> 1.37364, k2 -> 1.49254, n1 -> 0.147263, n2 -> 1.26496, \[CapitalDelta]1 -> 4, \[CapitalDelta]2 -> 1, e1 -> 1.01063, e2 -> 1.75675}} In[78]:= krit[1.6004404664797225, 1.4954588369276354, \ 0.010927309028896423, 1.2676565106600015, 4, 1, \ 0.010000289039330264, 1.7592900563124902, Fexp] Out[78]= 87.985 
Sort By:
Posted 10 years ago
 Maybe the intermediate values shown don't satisfy all constraints? It's pretty much impossible to say for sure without full code.
Community posts can be styled and formatted using the Markdown syntax.