Progress for week 48 (2014)
From Robin
(Difference between revisions)
(→Snorre) |
(→Stian) |
||
Line 40: | Line 40: | ||
=== Accounting === | === Accounting === | ||
- | + | * Worked on a program for easier visualization of results. | |
+ | * Ran tests on unfair evaluation in respect to the runtime. | ||
+ | ** Rewrote evaluation for reduced runtime variation. | ||
+ | * Re-evaluated some previous results. | ||
== Snorre == | == Snorre == |
Revision as of 10:54, 3 December 2014
Contents |
Student template (copy this for your entry)
Budget
- Todo 1
- Todo 2
Accounting
- Done 1
- Done 2
Rune
Budget
- Add agents to scenario generator
- Improve path-following/paths
- Investigate dynamic avoidance
Accounting
André
Budget
- Print the tower, and the gear that suppose to be on the stepper
- Check if the angle is correct read out
- Look at som wheel bearings for the rotating tower
Else-Line
Budget
- Continue testing on different parameters
- Create more graphs for comparisons
Accounting
- Tested on different number of iterations
- Created corresponding graphs
Stian
Budget
- Continue working on an easier way to display the disparity results for each algorithm and their pareto fronts.
- Compare trainingset results with testset results.
- Also reevaluate single-threaded to reduce runtime bias.
- Further algorithm implementations.
Accounting
- Worked on a program for easier visualization of results.
- Ran tests on unfair evaluation in respect to the runtime.
- Rewrote evaluation for reduced runtime variation.
- Re-evaluated some previous results.
Snorre
Accounting
- Done experiments both with the simple GA and the NSGA-II, but the results were non-deterministic.
- Started investigating this problem by modifying the controller system.
- Worked on the presentation.
- Started planning the thesis.