Your skill will accomplish what the force of many cannot |
|
PerlMonks |
comment on |
( [id://3333]=superdoc: print w/replies, xml ) | Need Help?? |
Perhaps you might start with a chart showing the progress as well as all the metrics you can think of to plot. Keep a manual control over the continue/abort decision for now. As a human viewing it, you'll probably have a decent prediction of whether it is worth continuing. You can then pick the metrics and limits that match your predictions best, and use them to grind the remaining tests unattended. Or, go meta, and set a genetic algorithm free on deciding when to stop your processing. :) In reply to Re: Instrumenting a genetic algorithm.
by SuicideJunkie
|
|