Ecosyste.ms: OpenCollective

An open API service for software projects hosted on Open Collective.

github.com/GeomScale/volesti

Practical volume computation and sampling in high dimensions
https://github.com/GeomScale/volesti

We use std min_element and max_element iterators for the min and max positions and values of the window W.

7296fc51bfe0314329900bc1dd4d5698e4addd6d authored over 6 years ago by tolischal <[email protected]>
Make changes in arguments of get_first_gaussian() and get_next_gaussian(), we have moved the output arguments at the end of the argument list. We change all not returning functions to void in gaussian_annealing.h and in gaussian_samplers.h. We add comments for the variable k in get_next_gaussian and iterators for the fn vector. We have renamed its to iterators. We change rand_exp_range_coord() to return a double.

8bd0b0d807f43713bf34ba43a38dcfcec7aaf2c2 authored over 6 years ago by tolischal <[email protected]>
Implement Welford's algorithm for mean and variance. Use constants for max nuber of iterations and tolerance in gaussian_samplers.h and gaussian_annealing.h header files.

5a26917ee419ac5c54479e7668127000568b7d00 authored over 6 years ago by tolischal <[email protected]>
compute radius of gaussian ball walk before the sampler's call. Remove class point_on_Dsphere. We move functions point_in_Dsphere, point_on_Dsphere from gaussian_samplers.h to samplers.h and implement function get_direction() to avoid duplicated code. We use these functions for RDHR and Gausian Ball Walk and first random point in convex polytope.

00ff4a929b6b4a93f7ddf4290ff5d5d458821eb0 authored over 6 years ago by tolischal <[email protected]>
move computation of the first coordinate point outside the loops of the ratios in gaussian_annealing_volume. Use const bool for the computation of the first coord point.

2abf7da996b2a0a4631edb533fe2599b299d7745 authored over 6 years ago by Tolis <[email protected]>
merge develop into schedule_annealing. Override gaussian_next_point() in order to sample the first point for the coordinate directions HnR. Make adjustments in get_annealing_schedule() and in volume_gaussian_annealing() for the first point coordinate case.

34e54961fc97ce8d1728eac19352318d49a52122 authored over 6 years ago by Tolis <[email protected]>
Merge pull request #6 from TolisChal/use_eigen_matrix

Use eigen matrix for the H-polytope representation. Able to use float or long double.

6c3001e24ac837b27c37e229acfc78eb59e6dff3 authored over 6 years ago by Tolis Chalkis <[email protected]>
Make adjustments in vol_R.cpp in order to use NT number type as it is declared in volume.h. Now we can use float or long double in Rcpp interface as well.

63807344a4d1a65958c6330f568c631c38e86c5e authored over 6 years ago by tolischal <[email protected]>
Use double for the nikolic minimum enclosing ellipsoid implementation irrespective of the number type we use for the implementaion. We can use float or long double. We made adjustments in tests cpp files in order to use the template number type. We pass all the tests for both double and float.

1e3ef48dedbb21d0c54b7c70ee70d22a391d4033 authored over 6 years ago by tolischal <[email protected]>
Use template number type in BallIntersevtPolytope class and NT number type in volume.h header file and in vol.cpp file. We have to use template number type in nikolic implementation in order to use different number type than double. Pass all the test locally.

308cdc60c12d4ed44831d5e1ff1b1172a397e614 authored over 6 years ago by tolischal <[email protected]>
Override line_intersect_coord() function in polytope class in order to compute the first point of the CDHR and not to test in every step if this is the first point. We initialize min_plus to max double value and max_min to min double value to avoid checking in every iteration if they are seted for the first time. We add comments in Polytope class and we remove useles comments from samplers.h header file.

82cd96ecb0211dabeb38e3edb03ba7fa405f584c authored over 6 years ago by tolischal <[email protected]>
use template number type everywhere in H-polytope class. All tests passed locally.

54d7d75c5753dd32832cd94144260821416c5636 authored over 6 years ago by tolischal <[email protected]>
add template parameters in every function. Use iterators in volume.h and in gaussian_annealing.h. Compute total steps only in verbose mode.

145564885c119ad0e94c27694bf3fc6422814b9c authored over 6 years ago by tolischal <[email protected]>
Merge branch 'develop' into schedule_annealing

62bd4e8adcdb40df7adcf1f51380818640c76fb6 authored over 6 years ago by tolischal <[email protected]>
Merge branch 'develop' into v_polytopes

be755a0b6ec8d233cb1c2e3a4c74c45f4b2b7ef9 authored over 6 years ago by tolischal <[email protected]>
merge develop into use_eigen_matrix to get the adjustments from Rcpp interface

81624b942d004719e18b81375572d681026ebc57 authored over 6 years ago by tolischal <[email protected]>
use chebyshev_center() function in Rcpp interface. Chenge NAMESPACE file to the Rcpp::skeleton NAMESPACE default.

5be598e246dc8734032742eaae7509dc561974ca authored over 6 years ago by tolischal <[email protected]>
remove stdMatrix and stdCoeffs from the implementation. Now only eigen matrices and vectors are used everwhere in the implementation. We have to change reff() function (add it to todo list).

0beb1e44820c2430ce266f7abda047638d61c942 authored over 6 years ago by tolischal <[email protected]>
chenge hit and run implementations in order to use eigen matrices. We add seters and geters for the eigen matrices and use them in rounding function

bbcb8175292d5b88628ef79047c69e02f3299762 authored over 6 years ago by Tolis <[email protected]>
use eigen matrices for the rounding

401fea0520c1242605c08f58f8248702de8b28c5 authored over 6 years ago by Tolis <[email protected]>
use eigen matrix for the chebychev ball

3e9fa53427cdcbca3826d636e4130f62cf6226d2 authored over 6 years ago by Tolis <[email protected]>
merge develop into v_polytopes and fix conflicts

38010883db5c309b405b5e4b6b69cc89189ce5ce authored over 6 years ago by Tolis <[email protected]>
merge develop into schedule_annealing and use chebyshev_center() function in CV algo as well

ddfd9b07bc24a2f2867814e3c065134f0bf6b3fb authored over 6 years ago by Tolis <[email protected]>
Merge branch 'develop' into schedule_annealing

e0208eecd7517dfb4e89bef656c711cfb91d4c2d authored over 6 years ago by Tolis <[email protected]>
change chebyshev_center() function in polytope class. Now we have to call this function to compute the chebychev ball and this function calls the interface of lpSolve to compute the ball.

d7f74fbfbe7c8298545266c6a7e92993d8e94f41 authored over 6 years ago by Tolis <[email protected]>
Merge remote-tracking branch 'volume_approximation/gsoc18' into develop

f80f11ee985481e45b6f72c15d3bdb224a9f283c authored over 6 years ago by Tolis <[email protected]>
remove useless comments from the code

f16db5e3527ad528d44bb9bee6f2315fe134ee72 authored over 6 years ago by Tolis <[email protected]>
create a function in solve_lp.h header file that solves the LPs for the hit and run. In Vpolytope class we define function intersect_line() and intersect_line_coord(). For the chebychev ball of the Vpoly we add a check: We check if the selected simplex is full dimensional or not (we compute the rank of the matrix).

705efe680b833cdb13932d80536712660653ae87 authored over 6 years ago by Tolis <[email protected]>
fix the bug for the random selection of d+1 vertices for the chebychev ball in Vpolytope

ad0fd1bd190de7426a0f763cfcf7d4dd28e58cfe authored over 6 years ago by Tolis <[email protected]>
Fixing licence details

1b13775b5c21c2e2c0b80b0df1e864fc20d36472 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Merge branch 'develop' into schedule_annealing

21bbf4792922815ebe1cd9f7e5588bca461dfc23 authored over 6 years ago by Tolis <[email protected]>
create a Vpolytope class. We use the proposal work. We create chebyshev_ball() functions for booth H and V polytopes, ball_walk, new function for a random point in d sphere. We change the structure var to give boolean ball_walk, radius for ball_walk and chebychev center.

2f0c38adde2e84fdccfeb7c1b31d615b7bcb8516 authored over 6 years ago by Tolis <[email protected]>
merge cgal_remove branch after PR completed and fix conflicts in rounding.j, makevars, rounding_test.cpp, chebychev_test.cpp, cmakeLists.txt files

1b284e654d071fa43f450a70cb207a7b59626ed3 authored over 6 years ago by tolischal <[email protected]>
Merge pull request #2 from TolisChal/cgal_remove

Rcpp package for VolEsti without CGAL dependencies.

9a31d85453481d38e4d270d4b1a9b42c32f6de5a authored over 6 years ago by Tolis Chalkis <[email protected]>
remove some useless comments in vol_R.cpp and change the inputs of rouning_min_ellips() for the chebychev ball

343ef50504dacfaf2388ffe1dd260e50aceee340 authored over 6 years ago by tolischal <[email protected]>
fix a bug in rounding test and allow more iterations

f5ff03152708823db53ed8039f57e7b10d7dbe07 authored over 6 years ago by tolischal <[email protected]>
update the pdf for the documentation. Add flag -DBOOST_NO_AUTO_PTR both in makevars and cmakelists.txt. Change the solveLP inputs, now it takes only the polytope and computes the chebychev ball. Fix the bug in readme for the command is used for the documentation.

a875e897bce8e4345c7e225fe0698c2ed2d04bac authored over 6 years ago by tolischal <[email protected]>
improve rounding test for skinny cubes. Round until ratio r/R drops or max_iters=10

d3178c62c29e07c0ee747599b152a20553b15d8f authored over 6 years ago by tolischal <[email protected]>
add a pdf documentation and instructions in readme how to create it locally. We improve rounding test with skinny_cubes and rotated_skinny_cubes [needs more improvment] and improve the chebychev test. Remove the rotation test

d265bae6ba7b12234f898eecad9c01474c08505d authored over 6 years ago by tolischal <[email protected]>
change round test, adding volume test for skinny cubes and rotated skinny cubes. Fix the warnings for the Rd files from devtools::check().

1280527f8651ce18ceee1434c85b3d810a41ebee authored over 6 years ago by tolischal <[email protected]>
add @param description in line 152 in utilities.R file in order to fix the warning from roxygen2 and to add description for the input parameter (list)

4d18004eea1de323ab9782e81877c3cd7f1d6b75 authored over 6 years ago by tolischal <[email protected]>
add all the cpp files for the tests from cgal_remove branch

8b3aed4e1c1b8bfb387a4590ba8bd917c607d837 authored over 6 years ago by tolischal <[email protected]>
add std::isinf() and std::isnan() functions in tests cpp files.

b366c9535de8815a29834ae8968d1cdaf8831f90 authored over 6 years ago by tolischal <[email protected]>
add tests for rotated skinny cubes. Create a cpp file in test folder.

279e617bf5347a79b94ac64cb770ab36490c4c05 authored over 6 years ago by tolischal <[email protected]>
create tests for the computation of chebychev ball and the rounding. We create two cpp files, one for each test, in test folder and we modify the cmakelists.txt

bc0a29712adc214b75f6e67160cf6889c238019a authored over 6 years ago by tolischal <[email protected]>
merge cgal_remove and fix conflicts

a139f0108bac8b9cc13cd5905631ea9d51a8dad9 authored over 6 years ago by Tolis <[email protected]>
fix mistakes from merging with schedule_annealing.

c6a6ea55bd9a0f869eb5ceb2520336de11167200 authored over 6 years ago by Tolis <[email protected]>
add boost files that are missing

b985f145e3800e8dc29c8d707b723b1f593363ce authored over 6 years ago by Tolis <[email protected]>
Write comments in roxygen style for the R functions and compile them in order to create .Rd files for the documentation

4227622b03e515f20e65dc4fd1160ed282efcf9d authored over 6 years ago by Tolis <[email protected]>
add tests for chebychev and comment out it because build with circleci fails. Needs debugging.

9caae705c9b70f3ce8c550814efca5b39473d80c authored over 6 years ago by tolischal <[email protected]>
add description for the CV algo in vol.cpp file when the user asks for help giving the -h or --help flag.

94467060a93c3abc4cd4dc14c1e5a30a98e1c9d1 authored over 6 years ago by tolischal <[email protected]>
Complete the parametrization of CV algo. The CV volume function takes as input the new structure. New flags are seted in order to give the CV input.

570aa68a3f84e9c9fb9af45d1030d9f8283a3a02 authored over 6 years ago by tolischal <[email protected]>
create a structure vars_g in volume.h header file for the gaussian annealing inputs.

824a5d0bef5ce1f8d2e9df42641713a4e565d46d authored over 6 years ago by tolischal <[email protected]>
add gaussian ball walk for sampling

680252019029c172ae299cc1638e1cd8e98d3266 authored over 6 years ago by tolischal <[email protected]>
Merge branch 'cgal_remove' into schedule_annealing

13435ed040f9ee39d6f01f986e76525c435156f0 authored over 6 years ago by tolischal <[email protected]>
Remove math format from README

d1e1e491dfed8d56f25cd84f87cb4f704653d3b8 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Math style changes in README

d2bca2826262e5f347cdb4b03b3b51693640551c authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Minor style changes in README

461e40cba606e52a6f73e1f598baf4ba16998970 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Change info and style in Readme

9c45ca0a1aa2e3622472a39783dcf6e0654e7c16 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Add circleci badge

def0b167bfa85e5b5d76e4a7cf988887f5d39c46 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Merge branch 'cgal_remove' into schedule_annealing

e835eda8dd93550937ac0f2f8bc0f63aec7d7da6 authored over 6 years ago by tolischal <[email protected]>
Delete .Rbuildignore file

152f5c8584350c0debdc441b39330a1e819bc379 authored over 6 years ago by Tolis Chalkis <[email protected]>
delete .gitignore file

40ea7870f439926f519bfb2d95557e761a914e24 authored over 6 years ago by Tolis Chalkis <[email protected]>
add boost files in external. Now it is able to be compiled without pre-install boost library in your system

9c79558b98c13ddfdf42fec2f1f1eb961b5d7e06 authored over 6 years ago by tolischal <[email protected]>
Create and enable tests verbose mode

75a5aef37b2b629d2a3b03e3c13f869b9cc2fafc authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
add lines at the end

b3eed143c5ff7c9a4466ac044cd65dd0a9b34359 authored over 6 years ago by tolischal <[email protected]>
resolving conflicts

9ce43845e11387471322c925734af8125eb2e89c authored over 6 years ago by tolischal <[email protected]>
resolving conflicts

6350c4325c262a23677827ee650159362512f1ec authored over 6 years ago by tolischal <[email protected]>
make the changes suggested by vissarion

fa79b046ba0054b84c0d90fa366e8177c87fc0c7 authored over 6 years ago by tolischal <[email protected]>
add function gaussian_next_point in order to use it when we have to sample just a point and to exploit coordinate direction better complexity.

b0b4b0539b816d12f0f60653c078e035ecaa02e9 authored over 6 years ago by tolischal <[email protected]>
implement coordinate direction hit and run

c9b1dd0f77869f75fa3db66e2832faa1fec27a05 authored over 6 years ago by tolischal <[email protected]>
fix a bug in rand_exp_range function and change the seed for the random sampler using chrono library. CV algo returns the correct result.

77c4689dca50499a5441031c830fb43ba6d66eef authored over 6 years ago by Tolis <[email protected]>
fix random direction sampler

1e7abb308faa4dda484023b8c765eebcb86627da authored over 6 years ago by tolischal <[email protected]>
merge cgal_remove branch into schedule_annealing

eaa06a64cc0e67f6f31ed38c3b7f8857be2f22ed authored over 6 years ago by tolischal <[email protected]>
Merge branch 'cgal_remove' into schedule_annealing

7ac3cedcfce64e09ee3e5561f8ac09aa95dfec86 authored over 6 years ago by tolischal <[email protected]>
Add tests using doctest library and CMake

61bfb0687f2b615af35906e525ef578624c26a4b authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
fix some errors in sampling functions while debbuging'

bc41b90fc934ceedc81a6aeb7984364e5b0d05e5 authored over 6 years ago by tolischal <[email protected]>
Remove error code return from help mode of volume

171b21c381ecdae9ce1dd679f2707d911d7ebb50 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
CircleCi2 script for compilation

0b1554a42c219faf05b04774d793a847e3254b5c authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
fix implementation of rand_exp_range() function for sampling from gaussian dist. Add some printing for debbuging.

38962b832d7d0890a9651cdc8ded508386436abd authored over 6 years ago by tolischal <[email protected]>
Merge remote-tracking branch 'tolis_fork/cgal_remove' into cgal_remove

d09d33d9066c55cf600972b8cd71b06a51516fc2 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
fix bugs and add some printing for verbose mode (and debbuging).Algo runs but returns wrong volume approximation.

b7de66f86e824c4f499cf31c626702839b0e777b authored over 6 years ago by Tolis <[email protected]>
Complete the implementation of volume approximation based on CV algo. The new function is volume_gaussian_annealing in volume.h header file. No tests have been made.

eb3d6734b829ea479ec88b4fcc8c4f1bbd8917c3 authored over 6 years ago by Tolis <[email protected]>
complete gaussian sampling with both random and coordinate directions hit and run. The samplers are implemented in gaussian_annealing.h header file.

9a75c8561d4ae7e830f1030c710b4655e9178074 authored over 6 years ago by Tolis <[email protected]>
Merge branch 'cgal_remove' into schedule_annealing

29179b774234a8ae78c07b695d0e7241d022470d authored over 6 years ago by Tolis <[email protected]>
change line_intersect() function in order to be compitable with gaussian sampling. Now it returns a pair of double and not points. We will merge this into schedule_annealing branch to continue working on cv algo.

fb23c2d3b0b0d183c8bb1d3e6ee67dbcbf1b73e7 authored over 6 years ago by Tolis <[email protected]>
add rand_exp_range function in samplers.h in order to sample from gaussian on the chord intersecting polytope in rdhnr.

c6ac4b8c8dc07839773d6bd6a6b3db6391db60c5 authored over 6 years ago by Tolis <[email protected]>
implement annealing_schedule in gaussian_annealing.h header file but without sampling.

75b5e5de297ccda34f22e686f9153a3f320657b9 authored over 6 years ago by Tolis <[email protected]>
merge cgal_remove into schedule_annealing

1b032bf0ddfc75bfa7e84e3f308dc573a44a6b2d authored over 6 years ago by Tolis <[email protected]>
Merge branch 'cgal_remove' into schedule_annealing

5760106b8318f8932c2f24f7dd482df7e4e0adac authored over 6 years ago by Tolis <[email protected]>
include boost header files we need in external folder. Improve linking between header files.

148129b766a67c06916a733612df1aefcd78aae7 authored over 6 years ago by tolischal <[email protected]>
Update config.yml

62433a9ea8454d95cad6627e80240daf73016830 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Merge remote-tracking branch 'origin/develop' into cgal_remove

0286059182aafb4ae12a3c7d7acdd4b42ed5253a authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
Create config.yml

c34e05c0b17f629b768edc442517eb33ac6dc455 authored over 6 years ago by Vissarion Fisikopoulos <[email protected]>
create folder for annealing, a header file and a function which gives the first gaussian

fa7b5968652192be1f08663f97023f6d3264f4f7 authored over 6 years ago by Tolis <[email protected]>
add copyright comments to lpsolve source files .c

caadbdffe2b2a6d27f41e92091ef4c6cccc467f3 authored over 6 years ago by tolischal <[email protected]>
add comments to lpsolve source files

f084248506beff4c0ef24078f74e0c90275a87ee authored over 6 years ago by tolischal <[email protected]>
improve readme

4ef05920f0efcb076194a845bf650b11d7b9d123 authored over 6 years ago by tolischal <[email protected]>
implement my mentor's comments on the PR except of these which refer to experiments

c702e0ff6ff98bc268c4cba79b8fa34a3f03bef0 authored over 6 years ago by tolischal <[email protected]>
compile lpsolve through Rcpp using only source files. Create lp_solve folder in /R-proj/src which contains all the .c files. We use makefile and source files from lpSolveAPI at github. For the c++ code we use liblpsolve55.so file for the moment.

17bdce5395117b0f2d0b0a3636f91ed4e3458c80 authored over 6 years ago by tolischal <[email protected]>