TimQu
|
fcd277c42a
|
added an option that enables building of state valuations. Also include the state valuations when the model is exported to .dot format
|
7 years ago |
|
TimQu
|
9039323fa9
|
optimized setting the epoch class
|
7 years ago |
dehnert
|
68f14c728a
|
added missing check for existence of model
|
7 years ago |
|
dehnert
|
b4a0016362
|
zero-reward MEC elimination for reachability rewards
|
7 years ago |
|
dehnert
|
fe8c3820fd
|
started cleanup of reachability rewards in sparse MDP helper
|
7 years ago |
|
dehnert
|
e5572db54e
|
eliminating ECs for sound value iteration for until probabilities
|
7 years ago |
|
dehnert
|
b4bfd0c39f
|
performance improvement in DS-MPI; some cleanups
|
7 years ago |
|
TimQu
|
630acb7459
|
parsing of until formulas with multiple bounds
|
7 years ago |
|
TimQu
|
36c3a4d9ef
|
Avoided conversion of memory states. They are now directly represented as 64 bit integers
|
7 years ago |
|
dehnert
|
19ac4a360f
|
intermediate commit
|
7 years ago |
|
TimQu
|
37d5dceaab
|
added small check
|
7 years ago |
|
TimQu
|
7fc65707a9
|
set the maximal value for each dimension. Also support for dependent dimensions
|
7 years ago |
|
TimQu
|
8f8f0bf804
|
corrected the search for the set of reachble epoch classes
|
7 years ago |
|
TimQu
|
d0209a804b
|
Setting reachable product states now also uses the transformMemoryState method
|
7 years ago |
|
dehnert
|
cb849a9ab8
|
started on computing upper bounds for rewards for interval value iteration
|
7 years ago |
|
dehnert
|
9d98bf5fa8
|
automatically switching solvers if soundness is enforced
|
7 years ago |
|
dehnert
|
df0b5fbfa5
|
fixed multiply-reduce operations in the presence of empty row groups
|
7 years ago |
|
TimQu
|
17a45a43eb
|
better transformation of memory states
|
7 years ago |
|
dehnert
|
d25cc4b05f
|
first version of sound value iteration
|
7 years ago |
|
TimQu
|
7d4a438e82
|
Fixes for lower bounds
|
7 years ago |
|
TimQu
|
ade8078759
|
added test for lower bounded properties
|
7 years ago |
|
dehnert
|
ec61e110f2
|
introducing solver formats to enable linear equation solvers to take the fixed point rather than the equation system formulation
|
7 years ago |
|
TimQu
|
b054b67312
|
first version for lower bounded properties
|
7 years ago |
|
TimQu
|
4ba20d11d4
|
more functionality for epoch manager
|
7 years ago |
|
dehnert
|
8e8fc34c30
|
fixed some TBB-related issues and added power method for linear equation systems
|
7 years ago |
|
dehnert
|
bac50a32ab
|
warkaround for gcc 7.2.0: make modernjson compile again
|
7 years ago |
|
dehnert
|
2d99ff3126
|
preserving action knowledge from first to second PRISM parser pass
|
7 years ago |
|
dehnert
|
c5134c364f
|
Extraction and update of TBB-parallelized stuff
|
7 years ago |
|
TimQu
|
d129683c61
|
gathered dimension related data into a struct. Also started with lower reward bounds
|
7 years ago |
|
TimQu
|
41cf4e76db
|
The solutions are now stored epoch-wise and will be erased as soon as all predecessor epochs are computed
|
7 years ago |
|
TimQu
|
3044aaa3f5
|
The product model is now handled in a separate class
|
7 years ago |
|
TimQu
|
1ccc241462
|
computations on epochs are now handled in a separate class
|
7 years ago |
|
TimQu
|
d3e50b8769
|
optimized productInState Computation
|
7 years ago |
|
dehnert
|
bba2832e5b
|
finished Walker-Chae method
|
7 years ago |
|
TimQu
|
47ab74a16b
|
implemented single objective queries
|
7 years ago |
|
TimQu
|
8b466f1fa7
|
extended multidimensional bounded until formulas to have different subformulas in each dimension
|
7 years ago |
|
dehnert
|
5440d164b2
|
started on Walker-Chae algorithm
|
7 years ago |
|
dehnert
|
c77b9ce404
|
gauss-seidel style multiplication for gmm++
|
7 years ago |
|
dehnert
|
43643b9699
|
bump gmm++ version to 5.2 (from 5.0)
|
7 years ago |
|
dehnert
|
d27954622a
|
slightly changed handling of gauss-seidel invocations in linear equation solver
|
7 years ago |
|
dehnert
|
6e548627ee
|
adding storm-pgcl as a dependency to target binaries
|
7 years ago |
|
dehnert
|
00f88ed452
|
gauss-seidel-style value iteration
|
7 years ago |
|
TimQu
|
cbe73081f8
|
zig zag epoch computation order
|
7 years ago |
|
TimQu
|
6ae05560fc
|
setting solutions in place
|
7 years ago |
|
dehnert
|
9d95d2adcf
|
first version of multiply-and-reduce (only for native)
|
7 years ago |
|
dehnert
|
e37d5869ef
|
extracted static version to separate cmake file
|
7 years ago |
|
dehnert
|
dd035f7f5e
|
allow for summand in matrix-vector multiplication
|
7 years ago |
|
TimQu
|
488479b5a7
|
better objectiveRewardFilter
|
7 years ago |
|
TimQu
|
cd6a79de23
|
used cache memory when checking each epoch
|
7 years ago |
|
TimQu
|
13ade58fa0
|
using pair<uint, uint> in the map instead of vectors of size 2
|
7 years ago |
