7b038db6d5
Fixed missing part for label preservation and added formula preservation check
6 years ago
31b50d76e9
clearer error message
6 years ago
3efee0d35d
changelog update: export of mtbdds
6 years ago
9ad8209a65
clarify that a formula needs to be added to do anything in storm-pomdp
6 years ago
a73c2691b6
Integration of the new settings in the DFT analysis
6 years ago
5aa19c9a58
Added settings for non-Markovian state elimination
6 years ago
88d6300084
Added option for label preservation to state elimination
6 years ago
f5b6bc84ba
circumvent problems with the bdd export
6 years ago
4f063cd233
tackling problems on unix
6 years ago
6ce429efc4
Added missing include
6 years ago
ec67166041
Decoupled preprocessing and SMT solving in commandline interface
6 years ago
449c513db2
Cleanup DFTASFChecker
6 years ago
75d28060cc
Moved failure bound computation to decouple it from the SMT checker
6 years ago
85e995c050
Merge branch 'master' of https://srv-i2.informatik.rwth-aachen.de/scm/git/storm
6 years ago
d295f6e777
export of bdds into dot and text format
6 years ago
9c74bbed24
Decoupled FDEP conflict search and SMT solver
6 years ago
ae5c001d24
Moved non-Markovian state eliminator to its own class
6 years ago
2709b7d828
Merge branch 'master' into dftFDEP
6 years ago
10bb42e0f6
First version of non-Markovian state elimination for MAs
6 years ago
9c52d0d2ab
Workaround for IntelTBB linker issue by CMake/Regex magic.
IntelTBB does not use symlinks (as commonly used) to reference its libraries but instead uses linker scripts.
These linker scripts do not work with GCC and linking fails.
As a workaround we manually set the correct library in CMake after extracting the path from the linker script with regex magic.
This workaround is highly hackish and might break in the future.
6 years ago
7fb660227f
Replaced assert(false) by throwing an exception
6 years ago
685b5c6b27
Throw exceptions after switch/case to silence compiler warnings about not returning anything
6 years ago
adfe82d0d6
Fixed typo to void
6 years ago
230a2c86d3
Mentioning QVBS as Benchmark source in README
6 years ago
549774abc9
Added state remapping in state elimination
6 years ago
d0596ecf0c
Changed cmake policy for finding packages: Now path specified via -D<Packagename>_ROOT=/path/to/package/ are automatically considered when searching for packages.
6 years ago
27e65d5669
Added construction of the state remapping for elimination of non-Markovian states in MAs
6 years ago
add2a40a62
Integrated results of FDEP conflict search in DFT state space generation
6 years ago
fedac853df
Fixed gitignore to only exclude build dirs in the root folder.
Otherwise, e.g. src/storm/builder/ would also be excluded.
6 years ago
b357868a32
GurobiLpSolver: Fixed rounding of integral results.
6 years ago
9780ffeb1e
Git ignore for any dir with infix 'build'
6 years ago
47344f9080
Removed unused flat_set includes
6 years ago
6a4c18e4a2
Use custom FlatSet to account for allocator changes in flat_set in Boost 1.70.
Boost 1.70 changed the default allocator parameter from new_allocator<T> to void to reduce symbol lenghts.
This reverts the default to the old allocator.
6 years ago
1e3686480a
is_equal_to_one() is not used in Boost 1.70
6 years ago
3616bdbf13
Added two test cases for the FDEP conflict search
6 years ago
e2ef6bc52a
Added missing initialization of result vector
6 years ago
589555c75f
Moved dynamic behavior computation from builder to DFT and added SEQ and SPARE cases
6 years ago
84467267e0
Second try to improve performance for relevant events
6 years ago
aa150fc2e3
Extended FDEP conflict search by not considering pairs of FDEPs with static behavior
6 years ago
bec75813b1
Added computation of dynamic behavior vector for DFTs
6 years ago
39ec751f8d
Removed debugging output
6 years ago
187d47d8ac
Try to improve performance for relevant events
6 years ago
9bfc7858d0
Added improved upper bound correction
6 years ago
583a880620
Adjusted DFT to SMT conversion to deal with constant failures
6 years ago
a0c42fa630
Added debugging messages for transformations
6 years ago
d0494d07e6
Get settings only once for takeFirstDependency
6 years ago
bc623d1203
MinMaxLinearEquationSolver: Added a flag 'hasNoEndComponent' that is true if the system is known to have no end components. This decides if policy iteration does require a valid initial scheduler.
Renamed the 'hasNoEndComponents' solver requirement to 'hasUniqueSolution' as this is the actual thing we require for, e.g. sound value iteration.
7 years ago
820b48354d
Silenced warning
7 years ago
24d0576009
Merge from master
7 years ago
9c28ed990e
Use isBasicElement() instead of type
7 years ago
Alexander Bork
Sebastian Junges
Matthias Volk
Tim Quatmann