adds testing suite for nested binary var matrix functions #2502

SteveBronder · 2021-06-09T23:47:27Z

Summary

So @andrjohns in #2461 you were totally correct that I wasn't handling nested array structures correctly for var matrix. This fixes that by

adding overloads for apply_scalar_binary() for handling var_value<Matrix> types
adding an overload for as_array_or_scalar() that makes a 2d array from std::vector<std::vector<Scalar>> types
Adding functions to the test suite for testing binary vectorized functions (also with integer containers on the left and right hand side)

With the overload for as_array_or_scalar() this was actually pretty darn easy!

Tests

@andrjohns can you check expect_ad_vectorized_matvar to make sure I'm catching all the specializations I need? I was a bit confused and wasn't totally sure if I caught them all. You can run the functions changed in the PR with

./runTests.py test/unit/math/mix/fun/bessel_first_kind_test.cpp \
test/unit/math/mix/fun/bessel_second_kind_test.cpp \
test/unit/math/mix/fun/beta_test.cpp \
test/unit/math/mix/fun/binary_log_loss_test.cpp \
test/unit/math/mix/fun/falling_factorial_test.cpp

Side Effects

Nope

Release notes

Adds nested vectorized functions for the new matrix type.

Checklist

Math issue How to add static matrix? #1805
Copyright holder: Steve Bronder

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)

stan-buildbot · 2021-06-10T08:23:46Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	1.97	1.95	1.01	1.29% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	1.0	0.47% faster
eight_schools/eight_schools.stan	0.07	0.06	1.02	1.5% faster
gp_regr/gp_regr.stan	0.1	0.1	1.02	1.63% faster
irt_2pl/irt_2pl.stan	3.13	3.1	1.01	0.86% faster
performance.compilation	64.34	60.8	1.06	5.5% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	6.2	6.08	1.02	1.87% faster
pkpd/one_comp_mm_elim_abs.stan	18.12	18.13	1.0	-0.01% slower
sir/sir.stan	81.74	78.91	1.04	3.47% faster
gp_regr/gen_gp_data.stan	0.02	0.02	1.03	2.78% faster
garch/garch.stan	0.33	0.31	1.05	4.96% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.25	0.26	0.99	-0.8% slower
arK/arK.stan	1.24	1.23	1.01	0.63% faster
arma/arma.stan	0.44	0.43	1.02	1.91% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.0	2.03	0.98	-1.53% slower
Mean result: 1.01698673147

Jenkins Console Log
Blue Ocean
Commit hash: 3fb1507

Machine information

ProductName: Mac OS X ProductVersion: 10.14.6 BuildVersion: 18G8022

CPU:
Intel(R) Core(TM) i7-8700B CPU @ 3.20GHz

G++:
Configured with: --prefix=/Library/Developer/CommandLineTools/usr --with-gxx-include-dir=/Library/Developer/CommandLineTools/SDKs/MacOSX10.14.sdk/usr/include/c++/4.2.1
Apple clang version 11.0.0 (clang-1100.0.33.17)
Target: x86_64-apple-darwin18.7.0
Thread model: posix
InstalledDir: /Library/Developer/CommandLineTools/usr/bin

Clang:
Apple clang version 11.0.0 (clang-1100.0.33.17)
Target: x86_64-apple-darwin18.7.0
Thread model: posix
InstalledDir: /Library/Developer/CommandLineTools/usr/bin

andrjohns · 2021-06-10T12:11:07Z

Great! I'll have a look on the weekend

andrjohns

Review (finally)! The testing all looks very comprehensive, so no issues there. Just a couple of suggestions for simplifying the apply_scalar_binary overloads

stan/math/rev/functor/apply_scalar_binary.hpp

andrjohns · 2021-06-20T14:10:54Z

stan/math/rev/functor/apply_scalar_binary.hpp

+ * Note: The return expresssion needs to be evaluated, otherwise the captured
+ *         function and scalar fall out of scope.
+ */


Can also cut this comment since unaryExpr doesn't get called for varmat types

stan/math/rev/functor/apply_scalar_binary.hpp

test/unit/math/mix/fun/beta2_test.cpp

test/unit/math/test_ad.hpp

test/unit/math/test_ad_matvar.hpp

…-binary-funcs1-varmat

…add is_container_or_var_matrix

…4.1 (tags/RELEASE_600/final)

stan-buildbot · 2021-06-23T03:35:04Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	3.1	3.15	0.98	-1.63% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.99	-0.77% slower
eight_schools/eight_schools.stan	0.12	0.11	1.06	5.45% faster
gp_regr/gp_regr.stan	0.16	0.16	0.98	-1.89% slower
irt_2pl/irt_2pl.stan	5.88	5.93	0.99	-0.97% slower
performance.compilation	88.99	86.84	1.02	2.41% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	9.19	8.74	1.05	4.88% faster
pkpd/one_comp_mm_elim_abs.stan	29.77	30.23	0.98	-1.54% slower
sir/sir.stan	124.56	125.1	1.0	-0.43% slower
gp_regr/gen_gp_data.stan	0.04	0.04	1.01	0.76% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	3.0	3.03	0.99	-0.96% slower
pkpd/sim_one_comp_mm_elim_abs.stan	0.39	0.4	0.99	-1.49% slower
arK/arK.stan	2.55	2.55	1.0	0.01% faster
arma/arma.stan	0.64	0.63	1.02	1.7% faster
garch/garch.stan	0.68	0.68	1.0	-0.02% slower
Mean result: 1.00419979892

Jenkins Console Log
Blue Ocean
Commit hash: 42b5287

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/math/rev/functor/apply_scalar_binary.hpp

andrjohns · 2021-06-25T11:44:06Z

stan/math/rev/functor/apply_scalar_binary.hpp

+template <typename T1, typename T2, typename F,
+          require_std_vector_vt<std::is_integral, T1>* = nullptr,
+          require_var_matrix_t<T2>* = nullptr>
+inline auto apply_scalar_binary(const T1& x, const T2& y, const F& f) {
+  check_matching_sizes("Binary function", "x", x, "y", y);
+  return f(x, y);
+}


Suggested change

template <typename T1, typename T2, typename F,

require_std_vector_vt<std::is_integral, T1>* = nullptr,

require_var_matrix_t<T2>* = nullptr>

inline auto apply_scalar_binary(const T1& x, const T2& y, const F& f) {

check_matching_sizes("Binary function", "x", x, "y", y);

return f(x, y);

}

template <typename T1, typename T2, typename F,

require_any_std_vector_vt<std::is_integral, T1, T2>* = nullptr,

require_any_var_matrix_t<T1, T2>* = nullptr>

inline auto apply_scalar_binary(const T1& x, const T2& y, const F& f) {

check_matching_sizes("Binary function", "x", x, "y", y);

return f(x, y);

}

Then you don't need the separate overloads

stan/math/rev/functor/apply_scalar_binary.hpp

…-binary-funcs1-varmat

…he inputs

…4.1 (tags/RELEASE_600/final)

stan-buildbot · 2021-06-29T21:37:01Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	3.11	3.06	1.02	1.75% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.96	-3.91% slower
eight_schools/eight_schools.stan	0.11	0.11	1.0	-0.15% slower
gp_regr/gp_regr.stan	0.16	0.16	0.99	-0.5% slower
irt_2pl/irt_2pl.stan	5.98	5.87	1.02	1.93% faster
performance.compilation	87.94	86.59	1.02	1.54% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	8.76	8.77	1.0	-0.16% slower
pkpd/one_comp_mm_elim_abs.stan	29.74	29.08	1.02	2.23% faster
sir/sir.stan	126.22	121.22	1.04	3.96% faster
gp_regr/gen_gp_data.stan	0.03	0.04	0.98	-1.85% slower
low_dim_gauss_mix/low_dim_gauss_mix.stan	3.07	3.0	1.02	2.19% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.42	0.39	1.09	8.09% faster
arK/arK.stan	2.56	2.55	1.0	0.44% faster
arma/arma.stan	0.63	0.63	1.0	-0.37% slower
garch/garch.stan	0.68	0.68	1.01	0.59% faster
Mean result: 1.01134889713

Jenkins Console Log
Blue Ocean
Commit hash: e23c84e

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

All looks good!

SteveBronder and others added 6 commits June 9, 2021 19:38

adds testing suite for nested binary var matrix functions

7e00c1a

Merge commit '134184699c81556c1cced08c7ca043e31072179e' into HEAD

b643173

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

f431f2a

…4.1 (tags/RELEASE_600/final)

fix docs and breakup beta tests for windows

86a783a

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

9f65ddf

…4.1 (tags/RELEASE_600/final)

fix include order error

3fb1507

andrjohns requested changes Jun 20, 2021

View reviewed changes

SteveBronder and others added 5 commits June 22, 2021 18:11

Merge remote-tracking branch 'origin/develop' into feature/vectorized…

d309731

…-binary-funcs1-varmat

remove unneeded overloads of var<matrix> for apply_scalar_binary and …

3848b10

…add is_container_or_var_matrix

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

5da0a5d

…4.1 (tags/RELEASE_600/final)

Update is_container_or_var_matrix.hpp

447e001

add back apply_scalar_binary for var<matrix> var<matrix> ops

42b5287

andrjohns requested changes Jun 25, 2021

View reviewed changes

SteveBronder and others added 4 commits June 29, 2021 10:28

Merge remote-tracking branch 'origin/develop' into feature/vectorized…

9dee467

…-binary-funcs1-varmat

reduces overloads for apply_scalar_binary with var matrix as one of t…

dfe705e

…he inputs

reduces overloads for apply_scalar_binary with var matrix as one of t…

6115bdd

…he inputs

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

e23c84e

…4.1 (tags/RELEASE_600/final)

andrjohns approved these changes Jun 30, 2021

View reviewed changes

SteveBronder merged commit 5c1dae4 into develop Jun 30, 2021

rok-cesnovar deleted the feature/vectorized-binary-funcs1-varmat branch June 30, 2021 06:10

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

adds testing suite for nested binary var matrix functions #2502

adds testing suite for nested binary var matrix functions #2502

SteveBronder commented Jun 9, 2021 •

edited

Loading

stan-buildbot commented Jun 10, 2021

andrjohns commented Jun 10, 2021

andrjohns left a comment

andrjohns Jun 20, 2021

stan-buildbot commented Jun 23, 2021

andrjohns Jun 25, 2021

stan-buildbot commented Jun 29, 2021

andrjohns left a comment

adds testing suite for nested binary var matrix functions #2502

adds testing suite for nested binary var matrix functions #2502

Conversation

SteveBronder commented Jun 9, 2021 • edited Loading

Summary

Tests

Side Effects

Release notes

Checklist

stan-buildbot commented Jun 10, 2021

andrjohns commented Jun 10, 2021

andrjohns left a comment

Choose a reason for hiding this comment

andrjohns Jun 20, 2021

Choose a reason for hiding this comment

stan-buildbot commented Jun 23, 2021

andrjohns Jun 25, 2021

Choose a reason for hiding this comment

stan-buildbot commented Jun 29, 2021

andrjohns left a comment

Choose a reason for hiding this comment

SteveBronder commented Jun 9, 2021 •

edited

Loading