 |
| excaliber[Free-DC.org] 2004-10-21 15:33:02 | First, would it be possible to graphically map the various "strains" of optimizations? If so, how would I best go about making such a mapper. Would it be possible to do this by comparing how similar the parameters to each optimization are, and then creating a final difference that sums the differences of the parameters? Or would that be highly inaccurate? I basically want to map the different strains by color and see how they are each progressing (and see if one is stalling, or another has good potential but not much work done, etc). Would be very interesting to see. Thanks |
| Stephen Brooks 2004-10-22 08:46:45 | There are a lot of ways of doing this. I've got a program around that projects the genome space into 3D (using principal component analysis) and that shows up strains pretty well. Another way that could be highly successful is by using a "classifier neural net", which is programmed to look for similar groups and colour each differently. |
