Generalize functions starting with s and t by t4c1 · Pull Request #1848 · stan-dev/math

t4c1 · 2020-04-22T09:31:29Z

Summary

Generalizes functions starting with s and t. All trace_*_quad_form functions are also optimized by replacing trace(multiply(A,B)) with transpose(A).cwiseProduct(B).sum().

Tests

This is just a refactor. No new tests.

Side Effects

None.

Release notes

Generalized functions starting with s and t.

Checklist

Math issue Generalize matrix function signatures #1470
Copyright holder: Tadej Ciglarič

The copyright holder is typically you or your assignee, such as a university or company. By submitting this pull request, the copyright holder is agreeing to the license the submitted work under the following licenses:
- Code: BSD 3-clause (https://opensource.org/licenses/BSD-3-Clause)
- Documentation: CC-BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
the basic tests are passing
- unit tests pass (to run, use: ./runTests.py test/unit)
- header checks pass, (make test-headers)
- dependencies checks pass, (make test-math-dependencies)
- docs build, (make doxygen)
- code passes the built in C++ standards checks (make cpplint)
the code is written in idiomatic C++ and changes are documented in the doxygen
the new changes are tested

…4.1 (tags/RELEASE_600/final)

…into generalize_fun_s_t

t4c1 · 2020-04-29T07:52:15Z

@serban-nicusor-toptal I think Jenkins has some problems with windows workers here.

t4c1 · 2020-04-29T07:53:13Z

Hmm, maybe its not limited to windows.

rok-cesnovar · 2020-04-29T08:19:33Z

Restart Stage sometimes fails to work, I restarted the entire job.

stan-buildbot · 2020-05-05T18:41:06Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	5.09	5.1	1.0	-0.15% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	1.0	0.29% faster
eight_schools/eight_schools.stan	0.09	0.09	1.01	0.95% faster
gp_regr/gp_regr.stan	0.23	0.23	0.99	-0.64% slower
irt_2pl/irt_2pl.stan	6.53	6.44	1.01	1.31% faster
performance.compilation	87.07	84.41	1.03	3.06% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.54	7.54	1.0	-0.03% slower
pkpd/one_comp_mm_elim_abs.stan	20.86	20.56	1.01	1.42% faster
sir/sir.stan	93.25	91.69	1.02	1.67% faster
gp_regr/gen_gp_data.stan	0.05	0.05	1.01	0.6% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.98	2.98	1.0	-0.2% slower
pkpd/sim_one_comp_mm_elim_abs.stan	0.31	0.33	0.93	-7.93% slower
arK/arK.stan	1.74	1.76	0.99	-1.03% slower
arma/arma.stan	0.77	0.76	1.01	1.04% faster
garch/garch.stan	0.57	0.57	1.0	-0.23% slower
Mean result: 1.00061473852

Jenkins Console Log
Blue Ocean
Commit hash: 39cf128

Machine information

ProductName: Mac OS X ProductVersion: 10.11.6 BuildVersion: 15G22010

CPU:
Intel(R) Xeon(R) CPU E5-1680 v2 @ 3.00GHz

G++:
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/usr/include/c++/4.2.1
Apple LLVM version 7.0.2 (clang-700.1.81)
Target: x86_64-apple-darwin15.6.0
Thread model: posix

Clang:
Apple LLVM version 7.0.2 (clang-700.1.81)
Target: x86_64-apple-darwin15.6.0
Thread model: posix

# Conflicts: # stan/math/prim/fun/mdivide_left_ldlt.hpp

t4c1 · 2020-05-08T09:17:07Z

@SteveBronder @andrjohns This one is also waitinig for a review.

andrjohns

Looks good, just some minor changes/clarifications

andrjohns · 2020-05-10T04:15:54Z

stan/math/prim/fun/scale_matrix_exp_multiply.hpp

 * @param[in] A Matrix
 * @param[in] B Matrix
 * @param[in] t double
 * @return exponential of At multiplies B


Similar doc error here

Similar to what? I don't understand.

Sorry, I meant the 'At multiplies B' instead of 'At multiplied by B'

andrjohns · 2020-05-10T04:19:34Z

stan/math/prim/fun/simplex_constrain.hpp

 * The transform is based on a centered stick-breaking process.
 *
- * @tparam T type of elements in the vector
+ * @tparam T type of the vector


stan/math/prim/fun/simplex_free.hpp

stan/math/prim/fun/singular_values.hpp

andrjohns · 2020-05-10T04:27:01Z

stan/math/prim/fun/size_mvt.hpp

 * @throw std::invalid_argument since the type is a scalar.
 */
-template <typename T>
+template <typename T, require_stan_scalar_t<T>* = nullptr>


Should probably use more informative template params (e.g. ScalarT,EigT) for consistency with the rest of the functions

stan/math/prim/fun/softmax.hpp

andrjohns · 2020-05-10T04:40:28Z

stan/math/prim/fun/to_array_1d.hpp

+  std::vector<T_val> result(matrix_size);
  for (int i = 0; i < matrix_size; i++) {
-    result[i] = datap[i];
+    result[i] = mat_ref.data()[i];


What about something like:

std::vector<T_val> result(mat_ref.data(), mat_ref.data() + mat_ref.size());

I just realized this wouldn't work for blocks (neither my code nor your suggestion)!

Added a test and fixed.

stan/math/prim/fun/to_matrix.hpp

andrjohns · 2020-05-10T05:01:06Z

stan/math/rev/fun/sd.hpp

+    using T_vi = promote_scalar_t<vari*, T_map>;
+    using T_d = promote_scalar_t<double, T_map>;
+    vari** varis
+        = reinterpret_cast<vari**>(ChainableStack::instance_->memalloc_.alloc(


ChainableStack::instance_->memalloc_.alloc_array<vari*> is preferred now

…4.1 (tags/RELEASE_600/final)

stan-buildbot · 2020-05-11T16:45:32Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	5.12	5.12	1.0	0.06% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.98	-2.33% slower
eight_schools/eight_schools.stan	0.09	0.09	1.01	0.65% faster
gp_regr/gp_regr.stan	0.22	0.24	0.95	-5.43% slower
irt_2pl/irt_2pl.stan	6.45	6.51	0.99	-0.96% slower
performance.compilation	85.49	84.53	1.01	1.12% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.55	7.72	0.98	-2.14% slower
pkpd/one_comp_mm_elim_abs.stan	21.04	21.0	1.0	0.21% faster
sir/sir.stan	94.95	98.92	0.96	-4.17% slower
gp_regr/gen_gp_data.stan	0.05	0.05	1.02	1.54% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.98	3.0	1.0	-0.49% slower
pkpd/sim_one_comp_mm_elim_abs.stan	0.34	0.31	1.09	8.0% faster
arK/arK.stan	1.75	1.8	0.97	-2.74% slower
arma/arma.stan	0.76	0.8	0.95	-5.02% slower
garch/garch.stan	0.57	0.59	0.97	-3.17% slower
Mean result: 0.991218107125

Jenkins Console Log
Blue Ocean
Commit hash: cbdd188

Machine information

ProductName: Mac OS X ProductVersion: 10.11.6 BuildVersion: 15G22010

CPU:
Intel(R) Xeon(R) CPU E5-1680 v2 @ 3.00GHz

G++:
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/usr/include/c++/4.2.1
Apple LLVM version 7.0.2 (clang-700.1.81)
Target: x86_64-apple-darwin15.6.0
Thread model: posix

Clang:
Apple LLVM version 7.0.2 (clang-700.1.81)
Target: x86_64-apple-darwin15.6.0
Thread model: posix

stan-buildbot · 2020-05-12T15:11:48Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	5.1	5.16	0.99	-1.22% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.97	-2.8% slower
eight_schools/eight_schools.stan	0.09	0.09	1.02	2.42% faster
gp_regr/gp_regr.stan	0.23	0.23	1.01	0.56% faster
irt_2pl/irt_2pl.stan	6.49	6.6	0.98	-1.77% slower
performance.compilation	85.5	84.41	1.01	1.28% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.62	7.72	0.99	-1.37% slower
pkpd/one_comp_mm_elim_abs.stan	21.02	20.92	1.0	0.49% faster
sir/sir.stan	91.03	94.11	0.97	-3.38% slower
gp_regr/gen_gp_data.stan	0.05	0.05	1.0	-0.16% slower
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.98	3.01	0.99	-1.23% slower
pkpd/sim_one_comp_mm_elim_abs.stan	0.32	0.33	0.95	-4.72% slower
arK/arK.stan	1.79	1.84	0.97	-3.07% slower
arma/arma.stan	0.8	0.83	0.96	-4.35% slower
garch/garch.stan	0.58	0.57	1.03	2.83% faster
Mean result: 0.989609436753

Jenkins Console Log
Blue Ocean
Commit hash: 72af837

Machine information

ProductName: Mac OS X ProductVersion: 10.11.6 BuildVersion: 15G22010

CPU:
Intel(R) Xeon(R) CPU E5-1680 v2 @ 3.00GHz

G++:
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/usr/include/c++/4.2.1
Apple LLVM version 7.0.2 (clang-700.1.81)
Target: x86_64-apple-darwin15.6.0
Thread model: posix

Clang:
Apple LLVM version 7.0.2 (clang-700.1.81)
Target: x86_64-apple-darwin15.6.0
Thread model: posix

andrjohns

LGTM

t4c1 and others added 6 commits April 22, 2020 11:26

generalized inputs to functions starting with s and t

8afc805

[Jenkins] auto-formatting by clang-format version 6.0.0-1ubuntu2~16.0…

6758931

…4.1 (tags/RELEASE_600/final)

fixed headers

daf3d15

Merge branch 'generalize_fun_s_t' of https://github.com/bstatcomp/math …

11e3d94

…into generalize_fun_s_t

bugfixed to_* functions

2e10ede

bugfix stan_print

f448c13

reordered stan_print definitions

39cf128

Merge branch 'develop' into generalize_fun_s_t

1475ff7

# Conflicts: # stan/math/prim/fun/mdivide_left_ldlt.hpp

andrjohns requested changes May 10, 2020

View reviewed changes

t4c1 and others added 3 commits May 11, 2020 10:54

addressed review comments

96f7dbc

Merge commit '1f82629326f64f51897945ef398ad05b162d6cac' into HEAD

6e8161e

[Jenkins] auto-formatting by clang-format version 6.0.0-1ubuntu2~16.0…

cbdd188

…4.1 (tags/RELEASE_600/final)

t4c1 mentioned this pull request May 11, 2020

Let value_of and value_of_rec return expressions #1872

Merged

5 tasks

addressed remaining comments

72af837

andrjohns approved these changes May 13, 2020

View reviewed changes

t4c1 merged commit 3b2dbbd into stan-dev:develop May 13, 2020

t4c1 deleted the generalize_fun_s_t branch July 10, 2020 07:09

rok-cesnovar mentioned this pull request Jul 20, 2020

Release of Stan Math 3.3 #1974

Closed

Uh oh!

Conversation

t4c1 commented Apr 22, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Tests

Side Effects

Release notes

Checklist

Uh oh!

t4c1 commented Apr 29, 2020

Uh oh!

t4c1 commented Apr 29, 2020

Uh oh!

rok-cesnovar commented Apr 29, 2020

Uh oh!

stan-buildbot commented May 5, 2020

Uh oh!

t4c1 commented May 8, 2020

Uh oh!

andrjohns left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

stan-buildbot commented May 11, 2020

Uh oh!

stan-buildbot commented May 12, 2020

Uh oh!

andrjohns left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

t4c1 commented Apr 22, 2020 •

edited

Loading