Don't throw when computing the inverse of a size zero matrix #1701

mcol · 2020-02-11T18:49:21Z

Summary

This reverts the earlier decision to throw when computing the inverse of a size zero matrix, as discussed in #1689. We need to return an empty matrix ourselves because Eigen asserts when computing the inverse of an empty matrix. Fixes #1463. Fixes #1432.

Tests

Updated accordingly. There were also a couple of errors in those tests concerning the type of exception thrown.

Side Effects

None.

Checklist

Math issue inverse should have consistent errors on size 0 inputs #1463, inverse_spd should have consistent errors on size 0 #1432
Copyright holder: Marco Colombo

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

stan-buildbot · 2020-02-12T00:42:57Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	4.84	5.07	0.95	-4.78% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.99	-0.8% slower
eight_schools/eight_schools.stan	0.09	0.09	1.03	3.17% faster
gp_regr/gp_regr.stan	0.22	0.22	1.01	0.9% faster
irt_2pl/irt_2pl.stan	6.06	6.06	1.0	-0.03% slower
performance.compilation	89.76	87.1	1.03	2.96% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.48	7.49	1.0	-0.09% slower
pkpd/one_comp_mm_elim_abs.stan	21.33	21.17	1.01	0.75% faster
sir/sir.stan	89.74	98.18	0.91	-9.4% slower
gp_regr/gen_gp_data.stan	0.05	0.05	0.99	-1.35% slower
low_dim_gauss_mix/low_dim_gauss_mix.stan	3.01	3.0	1.0	0.3% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.34	0.32	1.07	6.89% faster
arK/arK.stan	1.75	1.74	1.0	0.32% faster
arma/arma.stan	0.8	0.79	1.01	0.94% faster
garch/garch.stan	0.59	0.59	1.0	-0.04% slower
Mean result: 1.00100474161

Jenkins Console Log
Blue Ocean
Commit hash: 2f875bd

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/fun/inverse.hpp

bob-carpenter · 2020-02-12T01:31:07Z

I approved because I don't think it's worth changing return style, but I'll let @mcol decide. Feel free to merge as is.

mcol · 2020-02-12T08:52:40Z

Since I didn't think it was worth fixing just that one case, I corrected that pattern everywhere.

stan-buildbot · 2020-02-12T14:34:11Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	4.9	4.92	1.0	-0.45% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.97	-3.06% slower
eight_schools/eight_schools.stan	0.09	0.09	0.98	-2.03% slower
gp_regr/gp_regr.stan	0.22	0.22	1.0	0.2% faster
irt_2pl/irt_2pl.stan	6.06	6.1	0.99	-0.65% slower
performance.compilation	88.72	87.46	1.01	1.42% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.52	7.47	1.01	0.65% faster
pkpd/one_comp_mm_elim_abs.stan	21.15	20.86	1.01	1.38% faster
sir/sir.stan	89.71	91.5	0.98	-2.0% slower
gp_regr/gen_gp_data.stan	0.04	0.05	0.99	-1.21% slower
low_dim_gauss_mix/low_dim_gauss_mix.stan	3.01	3.0	1.0	0.43% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.32	0.31	1.02	2.43% faster
arK/arK.stan	1.75	1.74	1.0	0.19% faster
arma/arma.stan	0.79	0.79	1.0	-0.37% slower
garch/garch.stan	0.59	0.59	1.0	0.33% faster
Mean result: 0.998370784699

Jenkins Console Log
Blue Ocean
Commit hash: 310f2e1

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

bob-carpenter

This looks really great. I'm loving the new { } return syntax.

stan-buildbot · 2020-02-13T06:11:57Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	4.9	4.85	1.01	0.99% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.98	-2.32% slower
eight_schools/eight_schools.stan	0.09	0.09	0.99	-0.85% slower
gp_regr/gp_regr.stan	0.22	0.22	0.98	-1.65% slower
irt_2pl/irt_2pl.stan	6.08	6.05	1.0	0.41% faster
performance.compilation	88.21	87.23	1.01	1.1% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.48	7.49	1.0	-0.24% slower
pkpd/one_comp_mm_elim_abs.stan	21.6	21.16	1.02	2.03% faster
sir/sir.stan	91.53	92.32	0.99	-0.87% slower
gp_regr/gen_gp_data.stan	0.05	0.05	1.02	1.92% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	3.03	2.99	1.01	1.13% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.34	0.32	1.08	7.03% faster
arK/arK.stan	1.75	1.73	1.01	0.64% faster
arma/arma.stan	0.79	0.79	1.0	-0.06% slower
garch/garch.stan	0.58	0.58	1.0	-0.32% slower
Mean result: 1.00646858951

Jenkins Console Log
Blue Ocean
Commit hash: 310f2e1

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 · 2020-02-13T10:37:06Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	4.99	4.88	1.02	2.21% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	0.94	-6.22% slower
eight_schools/eight_schools.stan	0.09	0.09	1.01	1.44% faster
gp_regr/gp_regr.stan	0.22	0.22	0.96	-4.41% slower
irt_2pl/irt_2pl.stan	6.07	6.06	1.0	0.19% faster
performance.compilation	88.3	87.14	1.01	1.31% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.48	7.47	1.0	0.15% faster
pkpd/one_comp_mm_elim_abs.stan	20.66	21.27	0.97	-2.96% slower
sir/sir.stan	89.94	90.86	0.99	-1.02% slower
gp_regr/gen_gp_data.stan	0.05	0.04	1.02	2.4% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	3.0	3.0	1.0	-0.05% slower
pkpd/sim_one_comp_mm_elim_abs.stan	0.32	0.32	0.99	-0.54% slower
arK/arK.stan	1.74	1.75	0.99	-0.6% slower
arma/arma.stan	0.8	0.79	1.01	0.86% faster
garch/garch.stan	0.58	0.58	1.0	0.19% faster
Mean result: 0.995832839626

Jenkins Console Log
Blue Ocean
Commit hash: 310f2e1

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 · 2020-02-13T15:41:06Z

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_pois_regr/gp_pois_regr.stan	4.88	4.88	1.0	0.05% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.02	0.02	1.01	1.22% faster
eight_schools/eight_schools.stan	0.09	0.09	0.96	-3.8% slower
gp_regr/gp_regr.stan	0.22	0.22	1.0	0.08% faster
irt_2pl/irt_2pl.stan	6.07	6.08	1.0	-0.04% slower
performance.compilation	88.1	87.08	1.01	1.16% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	7.48	7.48	1.0	-0.01% slower
pkpd/one_comp_mm_elim_abs.stan	20.72	21.26	0.97	-2.6% slower
sir/sir.stan	89.78	90.96	0.99	-1.31% slower
gp_regr/gen_gp_data.stan	0.05	0.05	0.98	-2.44% slower
low_dim_gauss_mix/low_dim_gauss_mix.stan	3.03	3.01	1.01	0.67% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.31	0.31	1.0	-0.03% slower
arK/arK.stan	1.74	1.73	1.01	0.88% faster
arma/arma.stan	0.79	0.8	0.98	-1.79% slower
garch/garch.stan	0.58	0.59	0.98	-1.64% slower
Mean result: 0.993865823573

Jenkins Console Log
Blue Ocean
Commit hash: 310f2e1

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

Don't throw when computing the inverse of a size zero matrix

2f875bd

bob-carpenter previously approved these changes Feb 12, 2020

View reviewed changes

stan/math/rev/fun/inverse.hpp Outdated Show resolved Hide resolved

Change return style for empty matrices

310f2e1

mcol dismissed bob-carpenter’s stale review via 310f2e1 February 12, 2020 08:52

bob-carpenter approved these changes Feb 12, 2020

View reviewed changes

mcol merged commit 5bb32be into develop Feb 13, 2020

mcol deleted the bugfix/1463-inverse-size-zero branch February 13, 2020 17:47

SteveBronder mentioned this pull request Apr 16, 2020

Stan Math 3.2 release #1826

Closed

bbbales2 mentioned this pull request Apr 20, 2020

Stanc3 release for Cmdstan 2.23 stan-dev/stanc3#498

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Don't throw when computing the inverse of a size zero matrix #1701

Don't throw when computing the inverse of a size zero matrix #1701

mcol commented Feb 11, 2020 •

edited

Loading

stan-buildbot commented Feb 12, 2020

bob-carpenter commented Feb 12, 2020

mcol commented Feb 12, 2020 •

edited

Loading

stan-buildbot commented Feb 12, 2020

bob-carpenter left a comment

stan-buildbot commented Feb 13, 2020

stan-buildbot commented Feb 13, 2020

stan-buildbot commented Feb 13, 2020

Don't throw when computing the inverse of a size zero matrix #1701

Don't throw when computing the inverse of a size zero matrix #1701

Conversation

mcol commented Feb 11, 2020 • edited Loading

Summary

Tests

Side Effects

Checklist

stan-buildbot commented Feb 12, 2020

bob-carpenter commented Feb 12, 2020

mcol commented Feb 12, 2020 • edited Loading

stan-buildbot commented Feb 12, 2020

bob-carpenter left a comment

Choose a reason for hiding this comment

stan-buildbot commented Feb 13, 2020

stan-buildbot commented Feb 13, 2020

stan-buildbot commented Feb 13, 2020

mcol commented Feb 11, 2020 •

edited

Loading

mcol commented Feb 12, 2020 •

edited

Loading