Feature #119
Feature #284: === Fit ===
Fitting: implement boostrapping fit strategy to deal with local minima during the fit
| Status: | Rejected | Start date: | 31 Oct 2012 | |
|---|---|---|---|---|
| Priority: | Low | Due date: | ||
| Assignee: | - | % Done: | 0% | |
| Category: | - | |||
| Target version: | - |
History
#1 Updated by pospelov almost 8 years ago
- Status changed from Backlog to Sprint
#2 Updated by pospelov almost 8 years ago
- Assignee set to pospelov
#3 Updated by pospelov almost 8 years ago
- Target version set to Sprint 7
#4 Updated by pospelov almost 8 years ago
- Target version deleted (
Sprint 7)
#5 Updated by pospelov almost 8 years ago
- Assignee deleted (
pospelov)
#6 Updated by herck almost 8 years ago
- Status changed from Sprint to Backlog
#7 Updated by wuttke over 7 years ago
- Parent task set to #284
#8 Updated by wuttke over 7 years ago
- Priority changed from Normal to Low
#9 Updated by pospelov over 6 years ago
- Status changed from Backlog to Rejected
We have implemented already the possibility to use mixture of stochastic (Genetic) and gradient descent algorithms.
We have implemented also the possibility to resample input real data.
Now one have to investigate in details objective functions.
This issue I suggest to drop, since it is more in the line of long term ideas.