Feature/issue 2778 add add diag

[drezap]

Submission Checklist

• Run unit tests: ./runTests.py src/test/unit
• Run cpplint: make cpplint
• Declare copyright holder and open-source license: see below

Summary

Add the add_diag function to the language.

Intended Effect

Less for loops.

How to Verify

make stan/src/test/test-models/good/function-signatures/math/matrix/add_diag
with command stan.

Side Effects

None.

Documentation

I'll open up an issue to add to docs when this is merged.

Copyright and Licensing

Please list the copyright holder for the work you are submitting (this will be you or your assignee, such as a university or company):
Copyright 2019 Andre Zapico

By submitting this pull request, the copyright holder is agreeing to 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/)

[stan-buildbot]

(stat_comp_benchmarks/benchmarks/gp_pois_regr/gp_pois_regr.stan, 1.01)
(stat_comp_benchmarks/benchmarks/low_dim_corr_gauss/low_dim_corr_gauss.stan, 0.94)
(stat_comp_benchmarks/benchmarks/irt_2pl/irt_2pl.stan, 1.0)
(stat_comp_benchmarks/benchmarks/pkpd/one_comp_mm_elim_abs.stan, 0.96)
(stat_comp_benchmarks/benchmarks/eight_schools/eight_schools.stan, 1.06)
(stat_comp_benchmarks/benchmarks/gp_regr/gp_regr.stan, 1.0)
(stat_comp_benchmarks/benchmarks/arK/arK.stan, 1.0)
(performance.compilation, 1.03)
(stat_comp_benchmarks/benchmarks/low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan, 1.01)
(stat_comp_benchmarks/benchmarks/low_dim_gauss_mix/low_dim_gauss_mix.stan, 1.01)
(stat_comp_benchmarks/benchmarks/sir/sir.stan, 1.0)
(stat_comp_benchmarks/benchmarks/pkpd/sim_one_comp_mm_elim_abs.stan, 1.02)
(stat_comp_benchmarks/benchmarks/garch/garch.stan, 1.04)
(stat_comp_benchmarks/benchmarks/gp_regr/gen_gp_data.stan, 1.01)
(stat_comp_benchmarks/benchmarks/arma/arma.stan, 0.98)
Result: 1.00406145551
Commit hash: 2f76d60

[bob-carpenter]

Thanks. The only thing this needs is a few more instantiations in the tests. The reason we require that is we want to make sure the math lib is doing the right thing in terms of being able to compile all possible signatures the language will throw at it. Parameters provide var and data provide double.

[bob-carpenter]

This file should have tests for all possible combinations of parameter and data for all possible signatures. That's a total of 3 signatures * 4 instantiations each (data/data, data/parameter, parameter/data, parameter/parameter).

[stan-buildbot]

(stat_comp_benchmarks/benchmarks/gp_pois_regr/gp_pois_regr.stan, 1.0)
(stat_comp_benchmarks/benchmarks/low_dim_corr_gauss/low_dim_corr_gauss.stan, 0.98)
(stat_comp_benchmarks/benchmarks/irt_2pl/irt_2pl.stan, 1.0)
(stat_comp_benchmarks/benchmarks/pkpd/one_comp_mm_elim_abs.stan, 0.97)
(stat_comp_benchmarks/benchmarks/eight_schools/eight_schools.stan, 0.97)
(stat_comp_benchmarks/benchmarks/gp_regr/gp_regr.stan, 1.0)
(stat_comp_benchmarks/benchmarks/arK/arK.stan, 1.25)
(performance.compilation, 1.03)
(stat_comp_benchmarks/benchmarks/low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan, 1.05)
(stat_comp_benchmarks/benchmarks/low_dim_gauss_mix/low_dim_gauss_mix.stan, 0.99)
(stat_comp_benchmarks/benchmarks/sir/sir.stan, 0.99)
(stat_comp_benchmarks/benchmarks/pkpd/sim_one_comp_mm_elim_abs.stan, 0.98)
(stat_comp_benchmarks/benchmarks/garch/garch.stan, 1.11)
(stat_comp_benchmarks/benchmarks/gp_regr/gen_gp_data.stan, 0.99)
(stat_comp_benchmarks/benchmarks/arma/arma.stan, 0.85)
Result: 1.01087269323
Commit hash: 5b0d90d

[bob-carpenter]

Can someone remind which is better, greater or less than one. And also, what's the range of variation we expect.

We don't want to induce 1% performance regressions on PRs, but how much of this is noise?

[wds15]

Numbers greater than 1 indicate a speed up...but smaller than 3% changes are probably noise anyway.

[drezap]

Is this ready?