Tim Quatmann
|
ee350ca384
|
Use same precision as BeliefValueType when dealing with triangulation resolutions.
|
5 years ago |
|
Tim Quatmann
|
92aa029bc5
|
Removed debug output
|
5 years ago |
|
Tim Quatmann
|
55c4408c6a
|
Storing the observation resolutions as a float so that we can increase the resolution more accurately with a non-integer factor
|
5 years ago |
|
Tim Quatmann
|
2f2a007896
|
Implemented 'guessing' of initial pomdp schedulers for multiple guesses
|
5 years ago |
|
Tim Quatmann
|
2500cc0cd2
|
Fixed computation of relative gap for special cases (in particular l=u=0)
|
5 years ago |
|
Tim Quatmann
|
896d409602
|
Implemented simple (but incomplete) check to display whether the belief MDP is finite.
|
5 years ago |
|
Tim Quatmann
|
1766bc385e
|
POMDP Approximation: Use relative gap
|
5 years ago |
|
Tim Quatmann
|
fcbce6052c
|
Fixed getting invalid bounds if we abort during the initial approximation step.
|
5 years ago |
|
Tim Quatmann
|
2ebb5e8383
|
Fixed detection of fixpoints.
|
5 years ago |
|
Tim Quatmann
|
703bdc4eb9
|
Changed strategy of the dynamic triangulation approach such that the number of "missed" probabilities is minimized
|
5 years ago |
|
Tim Quatmann
|
6ee2ed8550
|
Merge remote-tracking branch 'origin/master' into prism-pomdp
|
5 years ago |
|
Tim Quatmann
|
71c410a3be
|
Added settings to switch between different triangulation modes.
|
5 years ago |
|
Tim Quatmann
|
fa10087fba
|
Implemented triangulation in a more dynamic way.
|
5 years ago |
|
Tim Quatmann
|
cc5faee9c0
|
Fixed initial size threshold for over-approx.
|
5 years ago |
|
Tim Quatmann
|
2ac1c73076
|
Change default initial resolution to 3
|
5 years ago |
|
Tim Quatmann
|
ab95e7d08b
|
BeliefManager: organized stored beliefs in buckets (beliefs with the same observation belong in the same bucket)
|
5 years ago |
Sebastian Junges
|
e6f1c573c4
|
recent changes added
|
5 years ago |
|
Sebastian Junges
|
10253b25f2
|
removed xerces-c source from storm. If xerces-c is unavailable, storm will build everything as before, but storm-gspn will not be able to load gspns in XML format.
|
5 years ago |
|
Tim Quatmann
|
6f476ef079
|
belief exploration: Improved fixpoint detection for over-approx
|
5 years ago |
|
Tim Quatmann
|
ddec9ce740
|
ApproximatePomdpModelchecker: Fixed output a little.
|
5 years ago |
Matthias Volk
|
06941e7c48
|
Setting 'dft-statistics' prints information about intermediate approximation results
|
5 years ago |
|
Matthias Volk
|
1a1664e350
|
Updated .gitignore
|
5 years ago |
|
Matthias Volk
|
a61ea32aea
|
Fixed some GCC warnings
|
5 years ago |
|
Matthias Volk
|
d3c8093e0f
|
Removed unnecessary semicolons
|
5 years ago |
|
Matthias Volk
|
f45db73afe
|
Support coloured output for GCC
|
5 years ago |
|
Tim Quatmann
|
a187a299af
|
Merge branch 'master' into prism-pomdp
|
5 years ago |
|
Tim Quatmann
|
5a221acbd0
|
Multi-objective model checking: Fixed incorrect computations for some models with end components. (Github Issue #75)
|
5 years ago |
|
Tim Quatmann
|
08c60bcb3d
|
Added OVISolverSettings to storm-pomdp
|
5 years ago |
|
TimQu
|
7504f6f315
|
Improved statistics output for refinements, added detection of fixpoints
|
5 years ago |
|
Tim Quatmann
|
5a76f7355d
|
Fixed an issue with refinement of under-approximation
|
5 years ago |
|
Tim Quatmann
|
16ad9d3a83
|
fixed storm-pomdp output a little.
|
5 years ago |
|
Tim Quatmann
|
96309e1f11
|
Merge branch 'master' into prism-pomdp
|
5 years ago |
|
Tim Quatmann
|
4eed592811
|
--timeout now just sends a SIGALRM signal (which can be catched by the signal handler).
|
5 years ago |
|
Tim Quatmann
|
8b4595042e
|
Only do iteration output if the result bound improved. Handle integer overflows for the observation resolution.
|
5 years ago |
|
Tim Quatmann
|
43220759f4
|
Implemented a time limit for exploration.
|
5 years ago |
|
Tim Quatmann
|
e81b8f1622
|
BeliefManager: fixed a few assertion conditions
|
5 years ago |
|
Tim Quatmann
|
c91c98f2de
|
Pomdp: Fixing result output with exact numbers
|
5 years ago |
|
Tim Quatmann
|
1763f0c582
|
Making sure that we only store the best bounds found so far. Also added some output for the resulting values in each iteration.
|
5 years ago |
|
Tim Quatmann
|
fa624d2a20
|
Introduced new settings for controlling the refinement strategy and whether to produce only upper and/or lower bounds
|
5 years ago |
|
Tim Quatmann
|
2d94e77f2a
|
Only display the bound that was requested.
|
5 years ago |
|
Tim Quatmann
|
6dd50575f9
|
New 'belief-exploration' setting (replaces gridapproximation setting)
|
5 years ago |
|
Tim Quatmann
|
37490a8eca
|
Started to integrate new refinement options.
|
5 years ago |
|
Tim Quatmann
|
45f9c66602
|
Merge branch 'master' into prism-pomdp
|
5 years ago |
|
Tim Quatmann
|
a728c01322
|
BitVector: Fixed an issue with the move assignment operator. The 'other' BitVector was left in an invalid state.
|
5 years ago |
|
Tim Quatmann
|
75d792e987
|
Implemented refinement heuristic.
|
5 years ago |
|
Tim Quatmann
|
45832d3de3
|
BeliefMdpExplorer: Implemented extraction of optimal scheduler choices and reachable states under these choices
|
5 years ago |
|
Tim Quatmann
|
61215e4b24
|
Over-Approximation: Taking current values as new lower/upper bounds for next refinement step.
|
5 years ago |
|
Tim Quatmann
|
4ea452854f
|
Fixes for scoring observations
|
5 years ago |
|
Matthias Volk
|
7e1f5bf2ac
|
Fixed handling of constant BE in approximation
|
5 years ago |
|
Matthias Volk
|
49dac54e8b
|
Fixed typos
|
5 years ago |
