Sébastien Villemot
a7cc466285
Bytecode: fix bug in sparse matrix multiplication routines
...
Since those routines compute Aᵀ·B, the number of rows of the output is equal to
the number of columns of A.
2022-02-28 12:18:58 +01:00
Sébastien Villemot
8536b63827
Bytecode: remove superfluous debug information
2022-02-18 14:17:37 +01:00
Sébastien Villemot
972461c326
Bytecode: add debugging information about input arguments
2022-02-17 17:41:10 +01:00
Sébastien Villemot
a376d8c9fe
Fix steady state computation with bytecode+block and solve_algo ⩽ 4 or ⩾ 9
...
– Temporary terms were not correctly passed between blocks
– solve_algo ⩾ 9 was incorrectly passed through bytecode own’s solver instead
of through dynare_solve
2022-02-17 17:28:37 +01:00
Sébastien Villemot
ceb3ec2487
Bytecode: remove unused constructors
2022-02-15 16:54:41 +01:00
Sébastien Villemot
863fa2de1c
Bytecode: perfect foresight simulation of *linear* “Solve forward simple/complete” blocks was giving was results
2022-02-15 16:54:41 +01:00
Sébastien Villemot
803e3721ec
Bytecode: correctly handle M_.{endo_names,exo_names,param_names}, which are now cell arrays
2022-02-14 15:06:07 +01:00
Sébastien Villemot
92e82cfa0a
Bytecode: remove unused function prototype
2022-02-14 14:40:34 +01:00
Sébastien Villemot
188df7711d
Bytecode: fix simulation of blocks “Solve forward/backward complete” with stack_solve_algo={0,1,4}
...
The problem would also manifest at steady state computation with solve_algo=6.
It could crash or give incorrect results.
2022-02-11 12:00:25 +01:00
Sébastien Villemot
d74cb031d2
Bytecode: fix error message about convergence failure that can happen in Solve Backward Complete as well
2022-02-10 17:27:07 +01:00
Sébastien Villemot
cdb4f872b2
Bytecode: fix crash with stack_solve_algo=1 on “Solve forward/backward” blocks
2022-02-10 17:15:24 +01:00
Sébastien Villemot
ef6e4c7745
Bytecode: remove unused code
2022-02-10 16:58:39 +01:00
Sébastien Villemot
0caefa1593
Bytecode: exit gracefully when M_ does not exist
2022-02-09 14:32:25 +01:00
Sébastien Villemot
8d3e0587dc
k_order_mean MEX: check number of output arguments
...
Incidentally, this fixes a gfortran warning
2022-02-04 14:38:30 +01:00
Sébastien Villemot
964763ef4d
pthread Fortran interface: fix internal type for representing pthread_t structure
2022-02-04 12:33:19 +01:00
Sébastien Villemot
39f47dc957
local_state_space_iteration_k MEX now uses the Fortran code (instead of the C++ code)
...
Closes : #1802
2022-02-04 11:49:46 +01:00
NormannR
5f828e241f
Parallelizes local_state_space_iteration_fortran
2022-01-31 18:44:48 +01:00
Sébastien Villemot
0c68e7336b
perfect_foresight_problem MEX: error out properly when called MATLAB function triggers an exception
...
Closes : #1832
2022-01-03 18:11:31 +01:00
Sébastien Villemot
5f675eee18
Add erfc() primitive
2021-12-07 17:00:59 +01:00
Normann Rion
cf829fb28e
A few fixes for k-order welfare assesment in `evaluate_planner_objective`
...
As suggested in !1962
2021-12-07 12:19:30 +01:00
Sébastien Villemot
10698ba896
Fix building of source package
...
Bug introduced in 06fd42c19b
.
2021-11-26 19:15:27 +01:00
Sébastien Villemot
06fd42c19b
Drop the dynare_simul_ MEX in favour of the new Fortran k_order_simul MEX
2021-11-26 17:37:53 +01:00
NormannR
f889a25e86
Amend the k_order_welfare routine for it to return the adequate output variables
2021-11-24 17:16:03 +01:00
NormannR
45aad05670
Implements a Fortran routine `k_order_moment` to compute the k-order unconditional moment of the vector of endogenous variables using simulation
2021-11-24 17:09:12 +01:00
NormannR
cb5971d60f
Fix typo in local_state_space_iteration_fortran
2021-11-24 14:32:37 +01:00
Sébastien Villemot
ebec02c6e4
local_state_space_iteration_2 MEX: yet more fixes to comments
...
In particular, use ε instead of u in mathematical expressions for consistency
with the C++ variable name.
2021-10-22 17:57:29 +02:00
Sébastien Villemot
0be7e0a0d7
local_state_space_iteration_2 MEX: more fixes to comments
2021-10-21 16:34:57 +02:00
Sébastien Villemot
05ea09eee9
local_state_space_iteration_2 MEX: error out properly when trying to use with BLAS+MATLAB+parallelization
...
By the way, rename the C preprocessor symbol so that it is undefined by
default.
2021-10-20 15:50:34 +02:00
Sébastien Villemot
e3b1f9e79a
local_state_space_iteration_2 MEX: various fixes to comments
2021-10-20 15:39:58 +02:00
Sébastien Villemot
ce282dc29c
local_state_space_iteration_2 MEX: fix bug when there are more shocks that states
...
The code that computes ghx·yhat+ghu·u (both with and without pruning) was
making the implicit assumption that q⩽n, i.e. that the number of shocks is less
than or equal to the number of states. If q>n, it would try to read invalid
memory references in ghx and yhat, and would thus either crash or return dummy
results.
Closes : #1820
2021-10-14 16:18:17 +02:00
Sébastien Villemot
e722e908e5
k_order_simul+local_state_space_iteration_fortran MEX: fix gfortran warnings
2021-09-24 16:44:40 +02:00
Normann Rion
6475c52b3d
Implements k-order simulation routine in Fortran (Ref: #1802 )
...
`k_order_simul` may replace `dynare_simul_` in the long run
2021-09-24 09:11:15 +02:00
Normann Rion
a576bdd03f
Fixing the header doc and automake files for the local state space iteration Fortran MEX
2021-09-24 09:11:06 +02:00
Sébastien Villemot
7083f1d692
Merge branch 'local_state_space_it' into 'master'
...
Implements the Fortran replacement of local_state_space_iteration_k (Ref #1802 )
See merge request Dynare/dynare!1918
2021-09-22 15:54:17 +00:00
NormannR
cab65dabb5
Implements `local_state_space_iteration_fortran`, the Fortran replacement of `local_state_space_iteration_k` (Ref #1802 )
...
- Creates the library `libkordersim` with all the relevant Fortran routines to `folded_to_unfolded_dr` and `local_state_space_iteration_fortran`
- Implements `folded_to_unfolded_dr`, which converts folded decision rule matrices to their unfolded counterparts
2021-09-22 16:28:46 +02:00
Sébastien Villemot
271d80abee
Bytecode: release the lock on dynamic.bin when an exception occurs
...
Closes : #1815
2021-09-20 17:28:43 +02:00
Sébastien Villemot
67c205e0a4
Fortran 2008: add interface for mxCreateStructMatrix
...
Needed by Normann.
2021-09-14 17:05:07 +02:00
Stéphane Adjemian (Ryûk)
e72dde69d3
Add option for trust region algorithm (mex).
...
trust_region_initial_step_bound_factor is determinining thye initial
step bound. Default value is 100 (previous hard coded value was 1).
2021-07-23 19:44:16 +02:00
Johannes Pfeifer
66dcef2754
k_order_welfare.cc: remove gratuitous screen output
2021-07-22 15:12:59 +02:00
Normann Rion
3d27672c58
kth-order approximation of conditional welfare
...
Partially addresses issue #1680 :
- unconditional welfare resorts to dynare++ simulation tools, which shall be updated very soon
TO DO:
- implement a function computing kth-order approximation of simulated moments of y
2021-06-15 15:04:31 +02:00
Sébastien Villemot
a7203095db
A_times_B_kronecker_C MEX: don’t bother adding “only” clauses to “use” directive
2021-06-14 15:28:34 +02:00
Sébastien Villemot
766fff88f6
Use secure URL for link to GNU licenses
2021-06-09 17:35:05 +02:00
Sébastien Villemot
a40a9b079c
sparse_hessian_times_B_kronecker_C MEX: bugfix relating to number of rows of B and C
...
In practice however, it seems that this bug never hits, because B and C always
have the same number of rows when the MEX is called.
2021-06-08 17:58:40 +02:00
Sébastien Villemot
3bd3c78e0e
A_times_B_kronecker_C MEX: rewrite in Fortran
2021-06-04 12:58:39 +02:00
Sébastien Villemot
c4ca0ef0b2
Fortran MEX: remove useless return statements after calls to mexErrMsgTxt()
2021-06-04 12:58:39 +02:00
Sébastien Villemot
f90a8ab070
Fortran MEX: mark array pointers returned by MEX functions as contiguous
...
This can make a difference when the return value of those function is directly
passed to a BLAS/LAPACK function.
On the other hand, if the return value is first stored in a pointer variable,
then it seems necessary to explicitly say that this pointer is also contiguous.
2021-06-04 12:58:39 +02:00
Sébastien Villemot
dc2695a11f
mjdgges and block_trust_region MEX: optimise by marking some function arguments as contiguous
...
This avoids unnecessary array copies before calling BLAS/LAPACK functions.
2021-06-04 12:58:36 +02:00
Sébastien Villemot
3a2335975a
Fortran MEX: add missing “implicit none” in blas and lapack modules
2021-06-04 12:57:47 +02:00
Sébastien Villemot
45181cfd97
Fortran MEX: add interface for dgemm
2021-06-04 12:57:32 +02:00
Sébastien Villemot
c9eb6920c9
local_state_space_iteration_k MEX: fix bug in output
...
In its output, the MEX was returning values for all endogenous variables, but
it was used in a context where only the variables from oo_.dr.restrict_var_list
were expected (as is done with local_state_space_iteration_2 MEX).
This commit fixes this discrepancy, and also fixes the test that was checking
that both MEX are returning the same output.
Closes : #1768
2021-03-10 13:45:29 +01:00
Sébastien Villemot
1a0ee5f055
Bytecode / sparse transpose: bug fix
...
The Jc vector has n+1 elements, where n is the number of columns. Moreover, the
size passed to memset() was wrong.
2021-02-09 16:05:11 +01:00
Sébastien Villemot
c9f6d3a626
Bytecode: more simplifications and modernizations
2021-02-09 16:00:53 +01:00
Sébastien Villemot
4a1dbc5ca4
Bytecode: fix bug introduced in previous commit
...
map::at() does bound checking, and throws an exception if out-of-range,
contrary to [] operator which returns a new element.
2021-02-03 19:52:25 +01:00
Sébastien Villemot
4893f0e82c
Bytecode: various simplifications and modernizations
2021-02-03 18:10:31 +01:00
Sébastien Villemot
30cfb0b03c
Bytecode: bugfix in conditional forecast
2021-02-01 17:54:36 +01:00
Sébastien Villemot
967966876f
Bytecode: remove obsolete LINBCG stuff
2021-02-01 15:26:33 +01:00
Sébastien Villemot
743469bfd7
Bytecode: remove stuff for creating a standalone debugging executable
...
This code no longer compiles. And it’s possible to debug the MEX directly.
2021-02-01 15:26:33 +01:00
Sébastien Villemot
bb5f58f55f
Bytecode: remove useless constant redefinition
2021-02-01 15:26:33 +01:00
Sébastien Villemot
eb29dc003a
Bytecode: remove CUDA-specific stuff
2021-02-01 15:26:31 +01:00
Sébastien Villemot
5ef440ad9b
Bytecode: remove MSVC++ specific portions
2021-02-01 14:50:46 +01:00
Sébastien Villemot
f9f50ff19a
Bytecode: port to C++11 using keyword
...
Performed using modernize-use-using from clang-tidy.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-use-using.html
2021-02-01 14:50:46 +01:00
Sébastien Villemot
47357bb432
Bytecode: port to C++11 nullptr keyword
...
Performed using modernize-use-nullptr from clang-tidy.
Manual intervention to remove old-style C casts to nullptr.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-use-nullptr.html
2021-02-01 14:50:42 +01:00
Sébastien Villemot
a2186b06b8
Bytecode: port to C++11 syntax for default constructors
...
Performed using modernize-use-equals-default from clang-tidy.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-use-equals-default.html
2021-02-01 14:50:42 +01:00
Sébastien Villemot
1aa0286f6d
Bytecode: port to C++11 emplace_back() method on containers
...
Performed using modernize-use-emplace from clang-tidy.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-use-emplace.html
2021-02-01 14:50:42 +01:00
Sébastien Villemot
93c6cb1c3c
Bytecode: use bool literals
...
Performed using modernize-use-bool-literals from clang-tidy
https://clang.llvm.org/extra/clang-tidy/checks/modernize-use-bool-literals.html
2021-02-01 14:50:42 +01:00
Sébastien Villemot
1a4340cb80
Bytecode: port to C++11 auto keyword
...
Performed using modernize-use-auto from clang-tidy.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-use-auto.html
2021-02-01 14:50:39 +01:00
Sébastien Villemot
63e922f637
Bytecode: port some constructors to pass-by-value using C++11 move semantics
...
Performed using modernize-pass-by-value from clang-tidy.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-pass-by-value.html
2021-02-01 14:50:10 +01:00
Sébastien Villemot
7840d1396e
Bytecode: port to C++11 range-based for loops
...
Performed using modernize-loop-convert from clang-tidy.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-loop-convert.html
2021-02-01 14:50:07 +01:00
Sébastien Villemot
2f587cf5c3
Bytecode: use C++ headers
...
Performed using modernize-deprecated-headers from clang-tidy.
https://clang.llvm.org/extra/clang-tidy/checks/modernize-deprecated-headers.html
2021-02-01 12:28:50 +01:00
Sébastien Villemot
c050325e3d
Bytecode: C++17 modernization / convert the remaining enums to enum classes
2021-02-01 12:10:47 +01:00
Sébastien Villemot
bc053018a0
Remove risky steady state (unfinished) code
...
Incidently, remove the possibility of passing model derivatives as arguments to
the k_order_perturbation. That possibility was only used by the risky steady
state code.
Closes : #1338
2021-01-22 11:32:36 +01:00
Stéphane Adjemian (Odysseus)
0e1c9312e2
Merge remote-tracking branch 'Dynare/master' into enterprise
2021-01-11 11:06:55 +01:00
Sébastien Villemot
55e25bcb7f
perfect_foresight_problem MEX: check the return code of mexCallMATLAB()
2020-12-07 15:03:35 +01:00
Sébastien Villemot
4e16ffcda2
Fortran MEX interface: minor fix for interface to dgemv
2020-10-21 18:17:41 +02:00
Sébastien Villemot
ce5aa33a33
k-order DLL: no longer write journal file on-disk, unless options_.debug = true
...
Closes : #1735
2020-10-15 17:19:30 +02:00
Stéphane Adjemian (Charybdis)
e690c5de5a
Merge Enterprise/block_trust_region into enterprise.
2020-09-24 16:05:55 +02:00
Stéphane Adjemian (Charybdis)
17e6421ea3
Merged master into enterprise.
...
Fixed conflict in
- doc/manual/source/running-dynare.rst
2020-09-23 15:28:19 +02:00
Sébastien Villemot
73253a0620
Bytecode + Block decomposition: fix crash with purely forward variables
...
After simulating a block containing purely forward variables (thus of type
“evaluate backward”), the it_ variable of the evaluator would be left in an
inconsistent state (typically 0, which means that taking the value of a lagged
variable would lead to an invalid read).
By the way, fix a symmetric problem for backward blocks (which could
potentially create a invalid read for purely backward models).
Ref. #1727
2020-09-22 16:45:44 +02:00
Sébastien Villemot
8803bc1479
Bytecode: minor improvement to debugging information
2020-09-22 16:42:33 +02:00
Sébastien Villemot
865ab47fa9
Provide block_trust_region MEX under solve_algo 13 and 14
...
- block trust region solver now available under solve_algo=13
It is essentially the same as solve_algo=4, except that Jacobian by finite
difference is not handled. A test file is added for that case
- block trust region solver with shortcut for equations that can be evaluated
is now available under solve_algo=14 (in replacement of the pure-MATLAB solver)
Closes : Enterprise/dynare#3
2020-09-18 18:01:56 +02:00
Sébastien Villemot
7e21bf2a10
Block trust region MEX: use MATLAB’s dmperm for the Dulmage-Mendelsohn decomposition
...
It turns out that MATLAB’s implementation is significantly faster than my own
Fortran implementation.
2020-09-18 17:55:58 +02:00
Sébastien Villemot
35a162c6a6
Block trust region MEX: fix memory leak in MATLAB function closure
2020-09-18 17:55:58 +02:00
Sébastien Villemot
adf1fdb009
Block trust region MEX: add safety check for squareness of blocks
2020-09-18 17:55:58 +02:00
Sébastien Villemot
9430b4e9ca
Block trust region MEX: add debugging information
2020-09-18 17:55:57 +02:00
Stéphane Adjemian (Odysseus)
248ad18846
Merge remote-tracking branch 'Dynare/master' into enterprise
...
Fixed conflicts in:
- .gitlab-ci.yml
- matlab/dynare_config.m
- matlab/dynare_solve.m
- matlab/initvalf.m
- matlab/perfect-foresight-models/make_ex_.m
- matlab/perfect-foresight-models/perfect_foresight_setup.m
- mex/build/matlab/Makefile.am
- mex/build/matlab/configure.ac
- mex/build/octave/Makefile.am
- mex/build/octave/configure.ac
- mex/sources/Makefile.am
- preprocessor
- tests/Makefile.am
2020-09-17 14:53:32 +02:00
Sébastien Villemot
f6c2471eef
New MEX for solving discrete Lyapunov equations with doubling algorithm
...
This is a Fortran 2008 rewrite of disclyap_fast.m.
Closes : #1737
2020-07-30 16:34:35 +02:00
Sébastien Villemot
3f53a94981
Add dpotrf to Fortran 2008 BLAS/LAPACK bindings
2020-07-30 16:34:35 +02:00
Sébastien Villemot
ea18431230
Various improvements to mjdgges MEX
2020-07-30 16:34:33 +02:00
Sébastien Villemot
f4a31a0d1f
Various improvements to Fortran 2008 interface to MATLAB MEX functions
...
— add interface for more functions (cell, logical, struct)
— add new mexPrintf wrapper that trims and prints a newline
— functions that take indices of type mwIndex now 1-based indices
— improve the wrapper for mxArrayToString so that it returns a character scalar
2020-07-30 14:45:50 +02:00
Sébastien Villemot
49b35be90a
Preprocessor: various improvements to the use_dll mode
...
In particular, higher order derivatives are now returned as sparse matrices by
the static/dynamic files, instead of 3-column matrices (which was inconsistent
with the M-file mode).
2020-06-23 19:38:49 +02:00
Sébastien Villemot
31a9a5e9ab
Bytecode MEX: various fixes to debug mode
2020-05-13 12:22:20 +02:00
Sébastien Villemot
f4f296edd6
Preprocessor update
...
— various refactoring and bugfixes in block decomposition
— small bugfix for PAC models
2020-05-06 18:17:25 +02:00
Sébastien Villemot
6e0f104d7e
Bump minimal MATLAB version to R2014a
...
Ref. #1713
2020-04-07 16:32:32 +02:00
Sébastien Villemot
43ed7b2b69
Merge remote-tracking branch 'community/master' into enterprise
2020-03-27 16:28:41 +01:00
Sébastien Villemot
fbfff18378
Preprocessor: various refactorings of block decomposition
...
Adapt bytecode MEX accordingly.
2020-03-20 18:51:37 +01:00
Sébastien Villemot
144bdf34ba
k-order MEX: fix crash on Octave when higher-order derivatives have numerical zero
...
In Octave, when some values given to the sparse() function are numerically
zero, then the nzmax of the generated sparse matrix is shrinked accordingly;
while under MATLAB, the nzmax is the length of the vector of values, zeros
included.
The check at the top of
DynamicModelMFile::unpackSparseMatrixAndCopyIntoTwoDMatData() would then fail
under Octave if some higher-derivatives had an element which is symbolically
non-zero but numerically zero.
We therefore relax the check, and accordingly adapt the code that handles
numerical zeros.
This bug was uncovered by tests/pruning/AnSchorfheide_pruned_state_space.mod,
which was failing under Octave.
2020-02-18 12:18:39 +01:00
Sébastien Villemot
ae3f150c99
MS-SBVAR MEX: various simplifications and modernizations
2020-02-10 16:25:03 +01:00
Sébastien Villemot
5f9b9db31e
Build system: add missing Fortran 2008 files to source tarball
2020-01-16 16:54:47 +01:00
Sébastien Villemot
7e770f69e7
Remove workaround for errors in MEX files
...
Because at some point throwing exceptions from MEX files (with mexErrMsgTxt())
was not working under Windows 64-bit, we had designed a workaround to avoid
using exceptions.
Most MEX files were returning an error code as their first (or sometimes last)
argument, and that code would have to be checked from the MATLAB code.
Since this workaround is no longer needed, this commit removes it. As a
consequence, the interface of many MEX files is modified.
For some background, see https://www.dynare.org/pipermail/dev/2010-September/000895.html
2020-01-10 18:33:11 +01:00
Sébastien Villemot
89dc67eb46
Convert mjdgges MEX to Fortran 2008
2020-01-09 09:23:07 +01:00
Sébastien Villemot
c0c95e1ceb
Add Fortran 2008 interface for a subset of BLAS/LAPACK functions
2020-01-06 18:29:47 +01:00
Sébastien Villemot
ce9eca366d
Add Fortran 2008 interface for a subset of MEX functions
2020-01-06 18:29:47 +01:00
Sébastien Villemot
b901b7af86
Global reindentation of MEX sources
2019-12-20 16:30:24 +01:00
Sébastien Villemot
996bdd6c64
New local_state_space_iteration_k MEX, for nonlinear filters at k-order
...
It applies the approximated policy function to a set of particles, using
Dynare++ routines.
There is support for parallelization, using Dynare++ multithreading
model (itself based on C++11 threads; we don’t use OpenMP because it is
incompatible with MKL). For the time being, default to a single thread. This
should be later refined through empirical testing.
2019-12-20 11:31:56 +01:00
Sébastien Villemot
6a26926892
Add new block_trust_region MEX
...
This MEX solves nonlinear systems of equations using a trust region algorithm.
The problem is subdivided in smaller problems by doing a block
triangularisation of the Jacobian at the guess value, using the
Dulmage-Mendelsohn algorithm.
The interface of the MEX is simply:
[x, info] = block_trust_region(f, guess_value);
Where f is either a function handle or a string designating a function.
f must take one argument (the evaluation point), and return either one or two
arguments (the residuals and, optionally, the Jacobian).
On success, info=0; on failure, info=1.
2019-12-03 16:17:16 +01:00
Sébastien Villemot
93bd817ccb
Add Fortran 2008 interface for a subset of BLAS/LAPACK functions
2019-12-03 16:17:12 +01:00
Sébastien Villemot
ba466d22c8
Add Fortran 2008 interface for a subset of MEX functions
2019-12-03 16:17:08 +01:00
Sébastien Villemot
9e222734a8
Bytecode: fix logic in routine that substracts two sparse matrices
...
The logic of the dynSparseMatrix::Sparse_substract_SA_SB() routine was
incorrect.
In some cases, it would read past the last nonzero elements of the A matrix,
and consequently write past the number of allocated nonzero elements of the C
matrix.
This would lead to crashes and, probably, to wrong results under certain
circumstances.
Closes : #1652
2019-11-27 16:57:47 +01:00
Sébastien Villemot
42f172dec3
perfect_foresight_problem MEX: optimization for linear models
...
When the model is linear, there is no need to reevaluate the Jacobian for each
time period, since it is invariant.
Closes : #1662
2019-11-14 14:45:10 +01:00
Houtan Bastani
afcdf46e80
capitalize Cepremap and Matlab, uncapitalize DYNARE
2019-10-02 14:09:21 +02:00
Sébastien Villemot
6ff7d4c56c
C++17: use structured bindings instead of std::tie
2019-09-11 16:06:35 +02:00
Houtan Bastani
1501032a3f
replace use of old-style casts
2019-07-30 15:15:23 -04:00
Sébastien Villemot
4030169db9
k_order_perturbation MEX: number of threads is now configurable
2019-07-09 17:33:14 +02:00
Sébastien Villemot
5b6c3bb452
k_order_perturbation MEX: small simplification in handling of NameLists
2019-07-09 17:33:14 +02:00
Sébastien Villemot
aaaa897a2a
k_order_perturbation MEX: improve input handling and sanitization
2019-07-09 17:33:05 +02:00
Sébastien Villemot
f8af21819e
Simplify the interface to perfect_foresight_problem MEX
2019-07-09 14:33:17 +02:00
Sébastien Villemot
984c60acb9
Various improvements related to the MEX R2018a API
...
— Use MX_HAS_INTERLEAVED_COMPLEX for testing for the new API
— Use the typed access functions under the new API, with one exception: we do
not use mxGetDoubles() instead of mxGetPr(), because it would just complexif
the code without any value added
For more details, see:
https://fr.mathworks.com/help/matlab/matlab_external/upgrade-mex-files-to-use-interleaved-complex.html
2019-07-09 11:54:12 +02:00
Sébastien Villemot
f45738c09e
local_state_space_iteration_2 MEX: fix error message
2019-06-28 17:49:02 +02:00
Sébastien Villemot
67a49fc67b
k-order MEX: fix incorrect variable name in comment
2019-06-28 15:24:30 +02:00
Sébastien Villemot
ca1a1ae0f2
local_state_space_iterations_2 MEX: minor stylistic improvement
2019-06-27 17:34:10 +02:00
Sébastien Villemot
7d9b2a557b
perfect_foresight_problem MEX: number of threads is now configurable
2019-06-27 17:00:12 +02:00
Sébastien Villemot
9557061fa3
Add new num_procs MEX file for getting number of available processing units
2019-06-27 14:34:09 +02:00
Sébastien Villemot
e96f180c50
Implement mxIsScalar() for MATLAB < R2015a (8.5)
2019-06-25 20:31:13 +02:00
Sébastien Villemot
7c39b12b7b
Always compile MEX files with OpenMP (when they can take advantage of it)
...
As a consequence, remove the --enable-openmp option of the configure script.
2019-06-25 17:26:17 +02:00
Sébastien Villemot
82cef48eb4
A_times_B_kronecker_C MEX: remove the OpenMP codepath
...
Testing shows that it is slower than the BLAS path.
2019-06-25 17:26:14 +02:00
Sébastien Villemot
c6a09a65e9
kronecker MEX: improve comments using UTF-8
2019-06-25 14:34:18 +02:00
Sébastien Villemot
7347554db5
perfect_foresight_problem MEX: fix compilation under Windows
2019-06-25 11:14:01 +02:00
Sébastien Villemot
5b591fac42
New perfect_foresight_problem MEX file
...
It constructs the stacked residuals and jacobian of the perfect foresight
problem.
It is an almost perfect replacement for the perfect_foresight_problem.m
routine, while being much more efficient.
Note however that the DLL never return complex numbers (it instead puts NaNs at
the place where there would have been complex). This may create problems for
some MOD files; the algorithms will need to be adapted to use a more
line-search method.
2019-06-24 17:53:59 +02:00
Sébastien Villemot
ec05451d1a
Remove symmetric elements in 3rd model derivatives
2019-06-17 16:18:07 +02:00
Sébastien Villemot
1d22536bc5
Add back workaround for MKL with DGGES and VSL
...
The workaround was removed in de159c0480
, but it
is still needed for MATLAB R2009b.
2019-05-21 17:28:11 +02:00
Sébastien Villemot
faa3185bdf
local state space iterations 2 DLL: various modernizations and simplifications
2019-04-30 15:54:49 +02:00
Sébastien Villemot
1199d4abae
kronecker DLLs: various modernizations and simplifications
2019-04-30 15:54:48 +02:00
Sébastien Villemot
de159c0480
mjdgges DLL: convert to C++ (for consistency with other DLLs)
...
Various modernizations and simplifications.
Also remove a workaround for a LAPACK bug in DGGES (the VSL argument was
apparently referenced even though JOBVSL="N"). Hopefully the bug has been fixed
everywhere now.
2019-04-30 15:54:45 +02:00
Sébastien Villemot
b115b76e0d
Kalman steady state DLL: various modernizations and simplifications
2019-04-30 15:54:43 +02:00
Sébastien Villemot
80267f56e5
Block kalman filter DLL: various modernizations and simplifications
2019-04-30 15:54:40 +02:00
Sébastien Villemot
f5e2bfc30a
Ask GCC to warn about C-style casts (-Wold-style-cast)
...
Adapt the code accordingly.
2019-04-23 13:03:20 +02:00
Sébastien Villemot
76e7acab7c
block_kalman_filter DLL: fix header file extension
2019-04-23 09:33:28 +02:00
Sébastien Villemot
c9bdc2c885
Dynare++: simplify methods for getting number of columns and rows of a matrix
...
Previously there were GeneralMatrix::numRows() and TwoDMatrix::nrows() for doing
the same thing (and same for columns and Const versions).
Merge these two into GeneralMatrix::nrows().
2019-04-16 12:44:02 +02:00
Sébastien Villemot
be170f131c
k-order DLL: fix compilation error on Windows introduced by b1ba53ce0
...
Ref #217
2019-04-16 12:13:30 +02:00
Sébastien Villemot
57848930c6
Use Unicode copyright symbol (in UTF-8 encoding)
...
Do not use it in MATLAB source files, since unfortunately the MATLAB editor
does not support UTF-8.
2019-04-16 11:42:16 +02:00
Sébastien Villemot
e281f35213
Lift upper restriction on approximation order
...
Ref #217
2019-04-15 17:34:01 +02:00
Sébastien Villemot
b1ba53ce05
dynare_simul_ DLL: adapt for an arbitrary approximation order
...
The last input argument is now a struct containing matrices g_0, g_1,…
Typically one can pass oo_.dr for this argument.
Ref #217
2019-04-15 17:34:01 +02:00
Sébastien Villemot
68aa1ace8f
k-order DLL: in use_dll mode, get model derivatives from preprocessor at an arbitrary order
...
Ref #217
2019-04-15 17:34:01 +02:00
Sébastien Villemot
a101457cf0
k-order DLL: fix wrong assertion in MATLAB mode
...
Bug introduced in 4c6e911d6
(only apparent under Octave, since under MATLAB
assertions are skipped).
Ref #217
2019-04-15 17:34:01 +02:00
Sébastien Villemot
2169341639
Move source code of gensylv and dynare_simul_ MEX to the same place as other MEX
2019-04-15 17:34:01 +02:00
Sébastien Villemot
4c6e911d69
k-order DLL: in MATLAB mode, get model derivatives from preprocessor at an arbitrary order
...
We no longer use the old ‘modfile.dynamic’ compatibility layer. We directly
call the ‘modfile.dynamic_g*’ functions.
Ref #217
2019-04-12 18:26:31 +02:00
Sébastien Villemot
c31ba8bebf
k-order DLL: in ‘use_dll’ mode, compute # of temp. terms. using M_.dynamic_tmp_nbr
...
The preprocessor no longer exports the ‘ntt’ symbol in the MEX.
Ref #217
2019-04-12 18:26:31 +02:00
Sébastien Villemot
d7dd7214c7
k-order DLL: convert model derivatives from Dynare to Dynare++ format at an arbitrary order
...
Ref #217
2019-04-12 18:26:30 +02:00
Sébastien Villemot
2dc988b69d
k-order DLL: make model derivatives storage ready for an arbitrary order
2019-04-10 19:00:49 +02:00
Sébastien Villemot
85f472c74b
Dynare++: new helpers for constructing {Const,}{TwoD,General}Matrix from mxArray*
2019-04-10 19:00:49 +02:00
Sébastien Villemot
d0b4429fb6
k-order DLL: misc minor improvements
2019-04-10 09:41:58 +02:00
Sébastien Villemot
b556290d60
k-order DLL: simplify and better document correspondence between Dynare and Dynare++ variable orderings
2019-04-10 09:23:32 +02:00
Sébastien Villemot
efa1f39e71
k-order DLL: finally adapt for M_.{endo,exo}_names as cell arrays
2019-04-08 18:47:49 +02:00
Sébastien Villemot
86a607a4fc
k-order DLL: make the output arguments ready for arbitrary order
2019-04-08 18:47:49 +02:00
Sébastien Villemot
fa64a25825
qmc_sequence DLL: various modernizations and simplifications
2019-04-03 19:04:36 +02:00
Sébastien Villemot
dd09264b03
k-order DLL: return unfolded matrices in 5th output argument
...
Thus we can remove unfolding code from k_order_pert.m.
2019-04-02 19:16:13 +02:00
Sébastien Villemot
20cbc30450
k-order DLL: various simplifications
2019-04-02 19:16:10 +02:00
Houtan Bastani
a93c6681c7
clang: can’t print non POD type through variadic function
2019-04-01 16:05:47 +02:00
Sébastien Villemot
4c11e9e9ec
Dynare++ kord and MEX: various modernizations and improvements
2019-03-07 19:46:15 +01:00
Sébastien Villemot
14f85cf448
Dynare++ kord library: various modernizations and improvements
...
Note that I removed several #define whose purpose was to avoid typing "typename
ctraits<t>::…". Even though this tends to complicates the code, this is
probably safer, especially since the #define was capturing a free variable (t).
2019-03-06 18:43:04 +01:00
Sébastien Villemot
6d99b6d120
Dynare++ kord library: various modernizations and simplifications
2019-03-05 18:35:35 +01:00
Sébastien Villemot
58bdee12c3
Dynare++ tensor library: refactor static stuff
...
- encapsulate it more clearly
- make it thread safe
- put back the Pascal triangle initialization stuff removed in 9512a32ad0
2019-02-21 19:00:50 +01:00
Sébastien Villemot
b72857d4ce
Dynare++: the get() method of tensor containers now return a reference
...
Previously it was returning a pointer, which is non-standard for containers,
and was leading to ugly dereferencing everywhere.
2019-02-20 18:06:58 +01:00
Sébastien Villemot
b5793ddd22
Dynare++: use std::unique_ptr for storage inside tensor containers
2019-02-20 17:04:34 +01:00
Sébastien Villemot
9dfb556740
k-order DLL: fix bug introduced in 1f7d3beddc
...
For the IntSequence class, the initialization with {} and () are no longer
equivalent.
2019-02-12 19:09:28 +01:00
Sébastien Villemot
1f7d3beddc
Dynare++ tensor library: modernize the Symmetry class
...
We now use a initializer list constructor for creating symmetries of the form
$y^n$, $y^n u^m$, $y^nu^m\sigma^k$.
The constructor taking a single integer is used to initialize a symmetry of a
given length.
Similar changes are made to IntSequence.
This behavior is similar to std::vector.
2019-02-11 16:45:07 +01:00
Sébastien Villemot
af722f438f
Dynare++: drop remaining "using namespace std"
2019-02-06 19:00:46 +01:00
Sébastien Villemot
733308807d
Dynare++: by default, use as many threads as there are logical CPUs
...
The former default was 2 threads, since Dynare++ was written at a time when
hyper-threading was being generalized (but multiple cores were not).
2019-01-29 16:34:25 +01:00
Sébastien Villemot
752a02a36c
Dynare++: use C++11 threads (instead of POSIX threads API)
...
On Windows, this means that a POSIX threads implementation is no longer needed,
since C++11 threads are implemented using native Windows threads.
On GNU/Linux and macOS, POSIX threads are still used under the hood.
A new m4 macro (AX_CXX11_THREAD) is used to add the proper compilation
flags (instead of AX_PTHREAD).
2019-01-28 18:44:24 +01:00
Sébastien Villemot
1ef1dedc65
Refactor GeneralMatrix and ConstGeneralMatrix classes
...
- Remove the GeneralMatrix(const ConstVector &) constructor, since it is hides
a memory allocation (copying the ConstVector into a fresh Vector). This
helped detecting and fixing several unneeded memory allocations. Some other
memory allocations are now more visible (with an explicit Vector{}
constructor).
- Add checks in GeneralMatrix(Vector, …) and ConstGeneralMatrix(ConstVector, …)
constructors for verifying that the {Const,}Vector has unit-stride (this was
an implicit assumption so far) and is large enough for storing rows*cols
elements.
- Add GeneralMatrix::operator=(const ConstGeneralMatrix &).
- Delete ConstGeneralMatrix::operator=().
2019-01-24 15:27:23 +01:00
Sébastien Villemot
c711d34d1d
Dynare++ / sylvester equation solver: refactor Vector and ConstVector classes
...
- these classes now encapsulate a std::shared_ptr<{const, }double>, so that
they do not perform memory management, and several {Const,}Vector instances
can transparently share the same underlying data
- make converting constructor from ConstVector to Vector explicit, since that
entails memory allocation (but the reverse conversion is almost costless, so
keep it implicit); do the same for GeneralMatrix/ConstGeneralMatrix,
TwoDMatrix/ConstTwoDMatrix
- remove the constructors that were extracting a row/column from a matrix, and
replace them by getRow() and getCol() methods on {Const,}GeneralMatrix
- rename and change the API of the complex version Vector::add(), so that it is
explicit that it deals with complex numbers
- add constructors that take a MATLAB mxArray
2019-01-22 16:13:22 +01:00
Sébastien Villemot
3c5afded8e
k-order DLL: fix build failure
2019-01-10 19:45:11 +01:00
Sébastien Villemot
188d6296d7
Dynare++: finish to apply Dynare C++ coding style and extensions (.cc/.hh)
2019-01-08 17:12:05 +01:00
Sébastien Villemot
ce1ef47093
dynare++ / tensor library (TL): move away from CWEB
...
By the way apply Dynare C++ coding style and extensions (.cc/.hh).
2019-01-08 16:09:25 +01:00
Sébastien Villemot
df772325a2
dynare++/kord: move away from CWEB
...
By the way apply Dynare C++ coding style and extensions (.cc/.hh).
2019-01-04 16:32:05 +01:00
Sébastien Villemot
dcac799289
Fix check for QZ eigenvalue criterium
...
The criterium was previously incorrectly applied to the square absolute value of
eigenvalues. Rather apply it to the absolute value itself (as done in
mjdgges.m and the AIM solver).
Closes #1632
2019-01-02 16:37:30 +01:00
Sébastien Villemot
331a27398a
Fix another memory leak in k-order DLL
...
Re Dynare/dynare#1490 .
2018-11-08 17:30:48 +01:00
Sébastien Villemot
50a09a687e
C++11 modernization
...
Incidentally, fix a memory leak related to DynamicModelAC instantiation, using
a std::unique_ptr (see Dynare/dynare#1490 ).
2018-11-08 17:24:19 +01:00
Sébastien Villemot
ffb36e5713
Remove obsolete kludges for MinGW
2018-10-02 17:10:36 +02:00
Sébastien Villemot
59642c3312
C output of preprocessor now splits generated fcts into several smaller fcts
...
Adapt the k-order DLL as a consequence.
2018-10-01 16:13:12 +02:00
Sébastien Villemot
d03cc56587
solve_algo=7 and stack_solve_algo=2 are now allowed under Octave
...
The gmres() function has been implemented (actually since Octave 3.4.1!).
2018-08-02 15:20:39 +02:00
Sébastien Villemot
932e5f946c
Update preprocessor submodule
...
Adapt bytecode DLL as a consequence.
2018-07-18 17:52:44 +02:00
Sébastien Villemot
cb46a2697e
Fix invalid memory write in k-order DLL
...
The bug would show at order 3 when the last output argument (derivs) is not
requested (in practice every 3rd order solution without pruning). The DLL would
still attempt to write into it, causing an invalid memory access.
2018-07-02 16:37:46 +02:00
Sébastien Villemot
b2c554623f
Various fixes related to the new layout of generated files
2018-06-28 18:14:40 +02:00
Sébastien Villemot
a1b8bd39b2
Move the location of various generated files on the filesystem
...
- M and MEX files are now under +${MODELNAME}/
- bytecode, C source and JSON now under ${MODELNAME}/model/
2018-06-27 17:03:39 +02:00
Sébastien Villemot
e1a3c40ef0
Fix compilation of MS-SBVAR DLL under Octave 4.4
2018-06-05 17:07:45 +02:00
Sébastien Villemot
02084ce310
Remove qzcomplex Oct-file
...
It's actually not needed. The QZ complex decomposition can be obtained in
Octave on real matrices by casting them to the complex type.
2018-05-17 12:11:15 +02:00
Sébastien Villemot
e2c2f2d966
Remove fallbacks for ilu, linsolve and ordschur under Octave
...
These functions are all included in Octave 4.2.
2018-05-16 10:48:44 +02:00
Sébastien Villemot
7a2aa211bf
Make mjdgges DLL compatible with MATLAB interleaved complex API
...
This API was introduced in MATLAB 9.4 (R2018a), because the internal
representation of complex numbers has changed.
2018-05-14 15:16:09 +02:00
Houtan Bastani
fb6a11f544
move estimation-dll to its own repository: https://github.com/DynareTeam/estimation-dll.git
2017-09-22 15:31:23 +02:00
Sébastien Villemot
1ff5c3f571
Fix various typos (thanks lintian).
2017-06-28 15:06:54 +02:00
Ferhat Mihoubi
119e5b4dfa
Corrects bugs in bytecode related to the use of finite instead of isfinite and to mxMalloc tests
2017-06-05 18:51:43 +02:00
Stéphane Adjemian (Charybdis)
1bf81c9f5a
Fixed copyright notices.
2017-05-18 18:36:38 +02:00
Stéphane Adjemian (Charybdis)
b4448937cc
Fixed indentation of c/h/cc/hh files.
2017-05-16 16:30:27 +02:00
Stéphane Adjemian (Charybdis)
a53636e24e
Fixed copyright notices.
2017-05-16 14:11:15 +02:00
Ferhat Mihoubi
981e31c8c3
Fix issues in bytecode with option extended_path related to the size of endo_simul and exo_simul
...
(cherry picked from commit ab7acc49e30fbcb292d63d58281581a308716b4c)
2017-05-08 18:40:18 +02:00
Ferhat Mihoubi
e248d8bf59
Fix memory and periods issues in bytecode
...
(cherry picked from commit fa5dd681abcb3243a5ee9be1f71fdfc51bced1cd)
2017-05-08 18:39:44 +02:00
Ferhat Mihoubi
857fc3c4f4
Corrects several bugs related to bytecode:
...
- Memory allocation is checked
- The amount of memory allocated for conditional forecast is corrected
2017-04-05 11:48:40 +02:00
Stéphane Adjemian (Charybdis)
1229372a6c
Defines M_PI (cross compilation for Octave).
2016-12-01 13:32:54 +01:00
Stéphane Adjemian (Charybdis)
23642b3b6b
Defines M_PI, M_SQRT2 and finite() in bytecode...
...
... Which are not in standard C/C++ (otherwise compilation of the mex for Octave
with -std=c++11 fails).
2016-12-01 13:32:54 +01:00
Houtan Bastani
e3fbefc616
mjdgges: fix bug where qz_criterium was not taken into account if a threshold argument is also passed
2016-10-27 11:19:34 +02:00
Houtan Bastani
8b17af05d5
bytecode: work around for isnan and isinf bug in C++ standard, glibc <= 2.23.
...
Fixed by https://sourceware.org/bugzilla/show_bug.cgi?id=19439
2016-10-11 16:01:27 +02:00
Michel Juillard
5d38280f27
remove one instance of dynamic exception specification
...
closes issue #1250
2016-07-16 00:13:47 +02:00
ferhat
ac2e5cfd52
- In extended path
...
- fix an error with results print out
- fix an error withe exogenous initialization
2015-11-19 17:03:05 +01:00
ferhat
77f5d572de
Fix the output arguments in case of bytecode failure
2015-11-06 15:06:17 +01:00
Stéphane Adjemian (Charybdis)
2ec1579cd0
Fixed warnings in bytecode compilation.
...
Removed unused variables.
2015-10-19 18:01:38 +02:00
ferhat
7c5272d852
Allows to compute the extended path and the conditional forecasts with bytecode
2015-09-22 15:39:49 +02:00
Houtan Bastani
1b869b26a7
fix dates for changes in 86c3c0fb1b
and a40323d2fb
2015-01-08 18:39:04 +01:00
ferhat
86c3c0fb1b
Checks that fields endo_names, exo_names and param_names exist
2015-01-06 10:30:52 +01:00
ferhat
24b2993f50
Solves issues in deterministic simulation:
...
- replaces maximum_endo_(lead|lag) by maximum_(lead|lag) to determine the maximum number of lead and lag in deterministic simulation
- allows to use bytecode in solve_perfect_foresight_model.m
- Adds model information in bytecode
2014-12-17 09:37:43 +01:00
Sébastien Villemot
41295ad602
Minor cleanups in k_order_perturbation DLL.
2014-03-21 17:26:55 +01:00
Sébastien Villemot
050b495a52
Fix an invalid memory access in k_order_perturbation DLL.
...
In the case where a 2nd/3rd derivative is symbolically not zero but numerically
zero at the evaluation point, the last lines of the g2/g3 matrices (in
KordpDynare::calcDerivativesAtSteady()) where uninitialized (these matrices
store the sparse hessian/3rd-deriv in coordinate list form, i.e. with 3 columns
and as many rows as non-zero elements). When reconstructing the dense
hessian/3rd-deriv matrix out of g2/g3, this would result in invalid memory
accesses.
2014-03-21 17:26:48 +01:00
Michel Juillard
45f7de33e8
sobol qmc_sequence MEX: dimension larger than 1111 returns a proper
...
Matlab error message instead of crashing
2014-02-15 21:11:42 +01:00
Sébastien Villemot
5308da2cad
Upgrade the minimum requirement to MATLAB 7.5 (R2007b).
2014-02-03 15:52:54 +01:00
Sébastien Villemot
e5d4defb41
Fix handling of qz_criterium in k_order_perturbation MEX.
...
If the MEX file was called with options_.qz_criterium=[], it would not use a
correct default value (but would use NaN).
Closes #574
2014-01-27 15:33:37 +01:00
Sébastien Villemot
66d48f5917
Compatibility fixes for Octave 3.8.
...
- no longer use OCTAVE_QUIT in MEX files; this seems now only possible in
oct-files. (Ref #304 )
- do not build linsolve.oct on Octave >= 3.8, it is available natively.
- do not add strjoin.m to the patch on Octave >= 3.8, it is available natively.
- default_save_options has been renamed save_default_options.
2014-01-03 11:05:39 +01:00
Sébastien Villemot
85f001885e
Estimation DLL: fix bug with creating MAT files
...
Similar problem to #446
2013-12-06 14:46:44 +01:00
Houtan Bastani
a8d07308a9
ms-sbvar: support ctrl-c changes in submodule, #304
2013-12-05 16:13:58 +01:00
Houtan Bastani
830c09760e
preprocessor: remove clang warning on declaration of const static double in class definition
2013-11-22 14:59:34 +01:00
Houtan Bastani
3c75923e2d
revert £ change as 156 is an extended ascii code
2013-11-05 19:06:20 +01:00
Houtan Bastani
0e4bb3f8e9
fix typo
2013-11-05 18:06:24 +01:00
Houtan Bastani
5352fb4d16
preprocessor and matlab mex fixes for clang compilation
2013-11-05 18:04:57 +01:00
ferhat
f4cee9d5e5
Corrects a bug in the way to test for the presence of the global variables
2013-10-28 14:26:42 +01:00
Stéphane Adjemian (Charybdis)
399b6d2615
Specialized maxit in bytecode mex (dynamic simulation vs steady state).
2013-10-09 16:55:46 +02:00
Stéphane Adjemian (Charybdis)
d8052342f3
Fixed bug introduced in cc0d9b42f0
(bytecode.cc).
...
Problem when reading nested structures.
2013-10-09 16:26:29 +02:00
Stéphane Adjemian (Charybdis)
cc0d9b42f0
Changed name of options_.maxit as options_.simul.maxit.
2013-10-09 13:06:06 +02:00
Sébastien Villemot
c34c90e8cb
Fix bug in linsolve when second argument is a zero-column matrix
...
Fixes crash of 1st-order purely forward models under Octave
2013-08-15 12:19:51 +02:00
Stéphane Adjemian (Charybdis)
5a4399e9b3
Removed openmp pragmas in bytecode.
...
This commit fixes #420 , so that bytecode doesn't crash when Dynare is compiled with --enable-openmp flag.
2013-07-11 15:12:41 +02:00
Stéphane Adjemian (Charybdis)
cb2228acad
Merge branch 'master' into experimental-mjdgges-threshold
2013-06-12 17:27:25 +02:00
Sébastien Villemot
61485ab809
Fix copyright notices
2013-06-12 17:04:46 +02:00
Sébastien Villemot
f1c8ec60c7
Remove various compilation warnings
2013-06-12 16:05:35 +02:00
Sébastien Villemot
555fc98673
Fix bug in gemv binding
2013-06-12 16:05:35 +02:00
Stéphane Adjemian (Charybdis)
7fc92c96d9
Fixed bug related to the test on the number of input arguments.
2013-06-12 10:21:27 +02:00
Stéphane Adjemian (Charybdis)
b81fcbeeb1
Added an option for the threshold level of the 0/0 generalized eigenvalue test.
...
The option is passed in the fourth input argument of mjdgges. If the
mex is called with less than four arguments, then the threshold level
takes its previous (hardcoded) default value: 1e-6.
2013-06-12 09:46:32 +02:00
Ferhat Mihoubi
b1da2f2b7d
"slowc" option is now available to solve a simple equation in bytecode
2013-06-10 16:46:54 +02:00
Ferhat Mihoubi
bdee6d14ec
Fix the problems related to the initialization of ilu preconditioner using BiCGStab and also a bug in the BiCGStab in case of block (without bytecode) model options.
2013-06-08 17:16:20 +02:00
ferhat
4e7050baf9
Use ilu with type=ilutp instead of nofill
...
Contrary to luinc command ilu with nofill option doesn't not allow for partial
pivoting
2013-05-23 11:44:29 +02:00
Sébastien Villemot
2670be408e
Bug fix in steady state solver
2013-04-19 18:01:19 +02:00
Sébastien Villemot
7495668ef2
Minor simplification and improvements
2013-04-19 18:01:19 +02:00
Sébastien Villemot
e1f702f83a
Various fixes to steady state computation in estimation DLL
...
Still not working… Valgrind detects unitialized values during the call to
hybridjs…
2013-04-18 18:28:34 +02:00
Stéphane Adjemian (Charybdis)
8eed899f92
Fixed bug. Compilation of bytecode was crashing with USE_OMP.
2013-04-18 11:20:35 +02:00
Sébastien Villemot
8dbf195422
Compatibility fix with GCC 4.8
2013-04-15 17:18:38 +02:00
Sébastien Villemot
3025a14ed9
Adapt for removal of luinc in MATLAB R2013a
2013-03-25 12:05:56 +01:00
Sébastien Villemot
37592309fb
Distribute dynumfpack.h in tarball
2013-03-24 09:42:45 +01:00
Sébastien Villemot
1db892efc4
Fix compilation problems of bytecode with UMFPACK and MATLAB's libut
2013-03-23 13:02:07 +01:00
Ferhat Mihoubi
19032f4b07
Adds an umfpack definition file for matlab on windows
2013-03-22 16:38:34 +01:00
Ferhat Mihoubi
43a50dd395
- adds the kernel CUDA file for SparseMatrix
2013-03-22 16:36:30 +01:00
Ferhat Mihoubi
dfa744fcb3
- adds new files: Evaluate.cc and Evaluate.hh to bytecode
2013-03-22 16:34:50 +01:00
Ferhat Mihoubi
2a51248832
Adds a block Kalman filter using GPU
2013-03-22 15:47:57 +01:00
Ferhat Mihoubi
03e487a092
Major update of bytecode:
...
- Iterative linear solvers using CUDA
- interpreter.cc decomposed
2013-03-22 15:44:34 +01:00