TimQu
|
17d6835477
|
removed acyclic minmax method as it is not much better then gauss-seidel style multiplications
|
7 years ago |
|
TimQu
|
d85a845b7d
|
Fixed cases where a solver guarantee was not established although it was needed by the termination condition
|
7 years ago |
|
TimQu
|
117d1b5c99
|
clean up single objective reward bounded case
|
7 years ago |
|
TimQu
|
016fedd58e
|
Better progress info
|
7 years ago |
dehnert
|
e6f61fb441
|
fixed bug in hybrid engine when using interval iteration
|
7 years ago |
|
TimQu
|
9859f60db0
|
Fixed solver tests
|
7 years ago |
|
TimQu
|
f44ce8801c
|
"fixed" the tests that failed because of unsound value iteration
|
7 years ago |
|
TimQu
|
3571f0ddca
|
Respected that the solution is unique when doing value iteration
|
7 years ago |
|
TimQu
|
33585c811f
|
MinMax Solver requirements now respect whether the solution is known to be unique or not.
|
7 years ago |
|
dehnert
|
4e4526182f
|
adding information about which technique is used to symbolic native linear equation solver
|
7 years ago |
|
dehnert
|
b8120ed73a
|
Markov automaton model checker now clearing basic requirements
|
7 years ago |
|
TimQu
|
3215f25513
|
upper result bounds for cumulative reward formulas to enable interval iteration
|
7 years ago |
|
TimQu
|
d91d979d90
|
changed some output
|
7 years ago |
|
dehnert
|
d90c507431
|
fixed bug in sparse bisimulation quotient extraction related to rewards
|
7 years ago |
|
TimQu
|
70dc9ce7ac
|
Bypassing requirements check to make value iteration without a lower result bound work
|
7 years ago |
|
TimQu
|
e3a506ecc6
|
Property information output
|
7 years ago |
|
dehnert
|
34485836b8
|
fixed some convertNumber applications
|
7 years ago |
Matthias Volk
|
f47e40d363
|
Fixed removed variable
|
7 years ago |
|
dehnert
|
4d7770aea6
|
fixed issue in hybrid reachability reward computation that caused empty row groups
|
7 years ago |
|
dehnert
|
3185719fe5
|
fix for linking issue under linux, adding missing l3pp dependency l3pp_ext
|
7 years ago |
|
dehnert
|
c0f07557ed
|
simplified state signature computation in dd-based bisimulation
|
7 years ago |
|
dehnert
|
29b915ccbf
|
fix out-of-index write in bisimulation quotienting
|
7 years ago |
|
TimQu
|
4cadc61f19
|
less searches for epoch solutions
|
7 years ago |
|
dehnert
|
0ef06fd31b
|
re-add time output to storm output and make iterative minmax solver respect linear equation solver format for policy iteration
|
7 years ago |
|
TimQu
|
b56cd07d0e
|
Consider only a submap of all epochsolutions for faster search
|
7 years ago |
|
TimQu
|
5c1de03d14
|
fixed min prob computation for single objective case
|
7 years ago |
|
dehnert
|
8204c03c0b
|
fixed a ton of warnings
|
7 years ago |
|
TimQu
|
5071df5c82
|
made sound value iteration work and respect the correct precision
|
7 years ago |
Sebastian Junges
|
ccfb1a292c
|
drn parser and exporter use reward names
|
7 years ago |
|
Sebastian Junges
|
2ce145745a
|
parameters from rewards are now also collected in wellformedness analysis
|
7 years ago |
|
Sebastian Junges
|
378e2ee40f
|
Better error message if expression parser fails
|
7 years ago |
|
TimQu
|
9140c1dc0e
|
statistics and empty-epoch-matrix optimization also for single objective case
|
7 years ago |
|
TimQu
|
108e8e69e8
|
Changed statistics output a little. Optimized the case where the transition matrix of the epoch model is empty
|
7 years ago |
|
Matthias Volk
|
0913388cd3
|
Renamed ExplicitDFTModelBuilderApprox to ExplicitDFTModelBuilder
|
7 years ago |
|
dehnert
|
8f95032d34
|
not logging exceeding precision in Kwek-Mehlhorn
|
7 years ago |
|
dehnert
|
bf727a28fd
|
remove debug output and choose sylvan automatically in exact mode
|
7 years ago |
|
TimQu
|
23686a0f09
|
reward bounded cumulative reward formulas + fixes for dimensions that do not need memory
|
7 years ago |
|
TimQu
|
fb4b5b2d84
|
added functionality to clear specified solution bounds of equation solvers
|
7 years ago |
|
TimQu
|
c7d9b0b61d
|
Fixed that bisimulation did not preserve reward bounded formulas
|
7 years ago |
|
Matthias Volk
|
87778a6775
|
Finally removed old DFTModelBuilder
|
7 years ago |
|
dehnert
|
1f16008b75
|
added proper exception handling to sylvan-based sharpening
|
7 years ago |
|
Matthias Volk
|
4456069e81
|
Fixed typo
|
7 years ago |
|
Matthias Volk
|
88544a9ec7
|
Added assertion for turnRatesToProbabilities in MA
|
7 years ago |
|
Matthias Volk
|
983e09fd63
|
Fixed possible problem with rates and exit rates in MA
|
7 years ago |
|
dehnert
|
d612337ebf
|
added bunch of debug output for aliasing problem
|
7 years ago |
|
dehnert
|
305f607507
|
fixed another parsing issue pointed out by Assistant Professor Dr. Nils Jansen
|
7 years ago |
|
dehnert
|
49cc789ca2
|
symbolic minmax solver respecting linear equation solver problem format; fixed parsing bug pointed out by Dr. Nils Jansen
|
7 years ago |
|
dehnert
|
a427eae699
|
fixed severe bug in symbolic bisimulation minimization
|
7 years ago |
|
Matthias Volk
|
91bd638c18
|
Fixed segfault
|
7 years ago |
|
dehnert
|
45a4b63a2e
|
fixed some issue in sylvan sharpen and forward minmax bounds to linear equation solver
|
7 years ago |
