jmdha FMA issues - Githubissues

jmdha / FMA

1 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Hiking does not find correct refinement

#116 kris701 opened 3 months ago
0
Rename `P10` repository to `FocusedMetaActions`

#115 kris701 opened 3 months ago
0
Document how to run this project on the cluster

#114 kris701 closed 3 months ago
0
Check driverlog for why the refinement is invalid

#113 jmdha closed 3 months ago
0
Meta action table

#112 jmdha closed 3 months ago
0
Replacement plan references unknown parameter

#111 jmdha closed 3 months ago
1
Save all CSV files from test runs in the repository in the end

#110 kris701 closed 3 months ago
0
Rewrite report to use plan length instead of search time

#109 kris701 closed 3 months ago
1
Mention in results that the usefulness-check based on time is a bit finiky

#108 kris701 closed 3 months ago
0
Add time limit for all candidates

#107 jmdha closed 3 months ago
0
Put all the final cluster scripts in the `Testing` folder

#106 kris701 closed 3 months ago
0
Well document how to run everything in the end

#105 kris701 closed 3 months ago
0
Run the testing side with the `Pham-Domains`

#104 kris701 closed 3 months ago
0
Rewrite the Usefulness Pruning section when better benchmarks are available

#103 kris701 closed 3 months ago
0
Test additional parameters

#102 jmdha closed 3 months ago
0
Fallback to "UsedInPlan" if the search time is too low during usefulness checks?

#101 kris701 closed 4 months ago
0
Ask alvaro to make a benchmark set based on the lama-first configuration?

#100 kris701 closed 3 months ago
0
Write out what search times during usefulness checks are

#99 kris701 closed 4 months ago
0
Make experiment setup section

#98 kris701 closed 4 months ago
0
Make Conclusion and Future Works section

#97 kris701 closed 3 months ago
0
Make a proper results section

#96 kris701 closed 3 months ago
0
Balance the dataset more?

#95 kris701 closed 4 months ago
0
Results of plan lengths with and without reconstruction

#94 kris701 closed 3 months ago
0
Sort learning benchmark problems by search time in the `properties` files

#93 kris701 closed 4 months ago
0
Add argument to change time limit for usefulness checks

#92 kris701 closed 4 months ago
0
Add a time limit for the caching process

#91 kris701 closed 4 months ago
0
Make a seperate stackelberg planner with the replacement plan output function

#90 kris701 closed 4 months ago
0
Setup tests to run with reconstruction

#89 kris701 closed 4 months ago
0
Setup meta action cache generation for reconstruction

#88 kris701 closed 4 months ago
0
Missing Learning Benchmarks

#87 kris701 closed 3 months ago
0
Domains that needs to be converted in the `Pham-Domains` folder

#86 kris701 closed 4 months ago
0
Reimplement the previous papers generation method, to get good quality benchmarks?

#85 kris701 closed 4 months ago
0
Extract experiment data from Alvaros previous paper to compare directly against?

#84 kris701 closed 4 months ago
0
Switch to use the same benchmarks as Alvaros previous paper

#83 kris701 closed 4 months ago
0
Ablation analysis in the results?

#82 kris701 closed 3 months ago
0
Deeper explenation of the stackelberg compilation in the background section

#81 kris701 closed 4 months ago
0
Make a planning background section

#80 kris701 closed 4 months ago
0
Make a "formal" problem definition section

#79 kris701 closed 4 months ago
3
Make candidate generation much more formal (include the concept of mutex groups)

#78 kris701 closed 4 months ago
0
Make a proposition, that fixing meta action candidates by mutex groups is always a good idea

#77 kris701 closed 4 months ago
0
Run experiments with 1 or 2 additional parameters

#76 kris701 closed 4 months ago
0
Split stackelberg into modified and original for stackelberg verfication

#75 jmdha closed 4 months ago
0
Rewrite introduction

#74 kris701 closed 4 months ago
0
Write problems remaining

#73 jmdha closed 4 months ago
0
Write the `Using the Meta Actions` section

#72 kris701 closed 4 months ago
0
Rewrite the `Refinement` section

#71 kris701 closed 4 months ago
0
Investigate `rovers`

#70 kris701 closed 4 months ago
0
Make section deeply discussing the refined meta actions

#69 kris701 closed 3 months ago
0
Make sure the refinement process can handle statics correctly

#68 kris701 closed 4 months ago
0
Document the CPDDL meta action generation

#67 kris701 closed 4 months ago
0