[integrals] fix substitution formula in meijerint #26173

maliesen · 2024-02-04T06:31:35Z

References to other Issues or PRs

Fixes #25786

Brief description of what is fixed or changed

fix substitution formula in meijerint

Other comments

Release Notes

integrals
- Fixed a bug with integrating over negative real numbers using meijerint algorithm. Formerly, integrate(exp(-x**2), (x,-5,0), meijerg=True) incorrectly gave -sqrt(pi)/2 * erf(5) instead of sqrt(pi)/2 * erf(5).

sympy-bot · 2024-02-04T06:31:42Z

✅

Hi, I am the SymPy bot. I'm here to help you write a release notes entry. Please read the guide on how to write release notes.

Your release notes are in good order.

Here is what the release notes will look like:

integrals
- Fixed a bug with integrating over negative real numbers using meijerint algorithm. Formerly, integrate(exp(-x**2), (x,-5,0), meijerg=True) incorrectly gave -sqrt(pi)/2 * erf(5) instead of sqrt(pi)/2 * erf(5). (#26173 by @maliesen)

This will be added to https://github.com/sympy/sympy/wiki/Release-Notes-for-1.13.

Click here to see the pull request description that was parsed.

<!-- Your title above should be a short description of what
was changed. Do not include the issue number in the title. -->

#### References to other Issues or PRs
<!-- If this pull request fixes an issue, write "Fixes #NNNN" in that exact
format, e.g. "Fixes #1234" (see
https://tinyurl.com/auto-closing for more information). Also, please
write a comment on that issue linking back to this pull request once it is
open. -->

Fixes #25786 

#### Brief description of what is fixed or changed

fix substitution formula in meijerint


#### Other comments


#### Release Notes

<!-- Write the release notes for this release below between the BEGIN and END
statements. The basic format is a bulleted list with the name of the subpackage
and the release note for this PR. For example:


* integrals
  * Fixed a bug with integrating over negative real numbers using meijerint algorithm. Formerly, `integrate(exp(-x**2), (x,-5,0), meijerg=True)` incorrectly gave `-log(x)`.

* physics.units
  * Corrected a semantical error in the conversion between volt and statvolt which
    reported the volt as being larger than the statvolt.

or if no release note(s) should be included use:

NO ENTRY

See https://github.com/sympy/sympy/wiki/Writing-Release-Notes for more
information on how to write release notes. The bot will check your release
notes automatically to see if they are formatted correctly. -->

<!-- BEGIN RELEASE NOTES -->
* integrals
  * Fixed a bug with integrating over negative real numbers using meijerint algorithm. Formerly, `integrate(exp(-x**2), (x,-5,0), meijerg=True)` incorrectly gave `-sqrt(pi)/2 * erf(5)` instead of `sqrt(pi)/2 * erf(5)`.
<!-- END RELEASE NOTES -->

Update

The release notes on the wiki have been updated.

oscarbenjamin · 2024-02-04T09:18:29Z

The tests that fails involves some integral of singularity functions that should probably not be handled by meijerg. Perhaps meijerg should reject such integrals to let another method take over.

github-actions · 2024-02-04T14:06:43Z

Benchmark results from GitHub Actions

Lower numbers are good, higher numbers are bad. A ratio less than 1
means a speed up and greater than 1 means a slowdown. Green lines
beginning with + are slowdowns (the PR is slower then master or
master is slower than the previous release). Red lines beginning
with - are speedups.

Significantly changed benchmark results (PR vs master)

Significantly changed benchmark results (master vs previous release)

| Change   | Before [a00718ba]    | After [d447356b]    |   Ratio | Benchmark (Parameter)                                                |
|----------|----------------------|---------------------|---------|----------------------------------------------------------------------|
| -        | 69.1±1ms             | 44.7±0.3ms          |    0.65 | integrate.TimeIntegrationRisch02.time_doit(10)                       |
| -        | 67.8±0.5ms           | 44.3±0.5ms          |    0.65 | integrate.TimeIntegrationRisch02.time_doit_risch(10)                 |
| +        | 18.6±0.4μs           | 30.9±0.5μs          |    1.67 | integrate.TimeIntegrationRisch03.time_doit(1)                        |
| -        | 5.35±0.03ms          | 2.84±0.01ms         |    0.53 | logic.LogicSuite.time_load_file                                      |
| -        | 73.1±0.6ms           | 28.9±0.1ms          |    0.4  | polys.TimeGCD_GaussInt.time_op(1, 'dense')                           |
| -        | 25.7±0.1ms           | 17.0±0.1ms          |    0.66 | polys.TimeGCD_GaussInt.time_op(1, 'expr')                            |
| -        | 73.3±0.3ms           | 29.2±0.2ms          |    0.4  | polys.TimeGCD_GaussInt.time_op(1, 'sparse')                          |
| -        | 256±0.8ms            | 125±0.2ms           |    0.49 | polys.TimeGCD_GaussInt.time_op(2, 'dense')                           |
| -        | 251±3ms              | 126±0.5ms           |    0.5  | polys.TimeGCD_GaussInt.time_op(2, 'sparse')                          |
| -        | 647±2ms              | 376±3ms             |    0.58 | polys.TimeGCD_GaussInt.time_op(3, 'dense')                           |
| -        | 651±1ms              | 375±2ms             |    0.58 | polys.TimeGCD_GaussInt.time_op(3, 'sparse')                          |
| -        | 497±1μs              | 289±1μs             |    0.58 | polys.TimeGCD_LinearDenseQuadraticGCD.time_op(1, 'dense')            |
| -        | 1.77±0.01ms          | 1.06±0ms            |    0.6  | polys.TimeGCD_LinearDenseQuadraticGCD.time_op(2, 'dense')            |
| -        | 5.86±0.02ms          | 3.16±0.03ms         |    0.54 | polys.TimeGCD_LinearDenseQuadraticGCD.time_op(3, 'dense')            |
| -        | 447±1μs              | 235±2μs             |    0.53 | polys.TimeGCD_QuadraticNonMonicGCD.time_op(1, 'dense')               |
| -        | 1.48±0.01ms          | 703±6μs             |    0.48 | polys.TimeGCD_QuadraticNonMonicGCD.time_op(2, 'dense')               |
| -        | 4.89±0.09ms          | 1.67±0.01ms         |    0.34 | polys.TimeGCD_QuadraticNonMonicGCD.time_op(3, 'dense')               |
| -        | 374±4μs              | 208±0.8μs           |    0.56 | polys.TimeGCD_SparseGCDHighDegree.time_op(1, 'dense')                |
| -        | 2.48±0.02ms          | 1.28±0.01ms         |    0.52 | polys.TimeGCD_SparseGCDHighDegree.time_op(3, 'dense')                |
| -        | 10.0±0.05ms          | 4.45±0.02ms         |    0.44 | polys.TimeGCD_SparseGCDHighDegree.time_op(5, 'dense')                |
| -        | 361±2μs              | 173±0.5μs           |    0.48 | polys.TimeGCD_SparseNonMonicQuadratic.time_op(1, 'dense')            |
| -        | 2.55±0.05ms          | 925±2μs             |    0.36 | polys.TimeGCD_SparseNonMonicQuadratic.time_op(3, 'dense')            |
| -        | 9.63±0.03ms          | 2.70±0.02ms         |    0.28 | polys.TimeGCD_SparseNonMonicQuadratic.time_op(5, 'dense')            |
| -        | 1.02±0.01ms          | 429±2μs             |    0.42 | polys.TimePREM_LinearDenseQuadraticGCD.time_op(3, 'dense')           |
| -        | 1.72±0.01ms          | 497±1μs             |    0.29 | polys.TimePREM_LinearDenseQuadraticGCD.time_op(3, 'sparse')          |
| -        | 5.85±0.04ms          | 1.86±0.01ms         |    0.32 | polys.TimePREM_LinearDenseQuadraticGCD.time_op(5, 'dense')           |
| -        | 8.40±0.06ms          | 1.48±0.01ms         |    0.18 | polys.TimePREM_LinearDenseQuadraticGCD.time_op(5, 'sparse')          |
| -        | 286±2μs              | 64.4±0.4μs          |    0.22 | polys.TimePREM_QuadraticNonMonicGCD.time_op(1, 'sparse')             |
| -        | 3.47±0.03ms          | 398±4μs             |    0.11 | polys.TimePREM_QuadraticNonMonicGCD.time_op(3, 'dense')              |
| -        | 4.03±0.03ms          | 277±0.5μs           |    0.07 | polys.TimePREM_QuadraticNonMonicGCD.time_op(3, 'sparse')             |
| -        | 6.94±0.04ms          | 1.30±0.01ms         |    0.19 | polys.TimePREM_QuadraticNonMonicGCD.time_op(5, 'dense')              |
| -        | 8.69±0.07ms          | 835±4μs             |    0.1  | polys.TimePREM_QuadraticNonMonicGCD.time_op(5, 'sparse')             |
| -        | 5.00±0.03ms          | 2.98±0.01ms         |    0.6  | polys.TimeSUBRESULTANTS_LinearDenseQuadraticGCD.time_op(2, 'sparse') |
| -        | 12.0±0.05ms          | 6.97±0.04ms         |    0.58 | polys.TimeSUBRESULTANTS_LinearDenseQuadraticGCD.time_op(3, 'dense')  |
| -        | 22.0±0.07ms          | 9.02±0.03ms         |    0.41 | polys.TimeSUBRESULTANTS_LinearDenseQuadraticGCD.time_op(3, 'sparse') |
| -        | 5.24±0.02ms          | 862±2μs             |    0.16 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(1, 'sparse')    |
| -        | 12.6±0.1ms           | 6.97±0.01ms         |    0.55 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(2, 'sparse')    |
| -        | 102±0.9ms            | 26.9±0.09ms         |    0.26 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(3, 'dense')     |
| -        | 166±1ms              | 53.6±0.1ms          |    0.32 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(3, 'sparse')    |
| -        | 174±0.9μs            | 114±0.6μs           |    0.65 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(1, 'dense')      |
| -        | 359±2μs              | 216±1μs             |    0.6  | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(1, 'sparse')     |
| -        | 4.28±0.02ms          | 877±4μs             |    0.21 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(3, 'dense')      |
| -        | 5.18±0.05ms          | 379±2μs             |    0.07 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(3, 'sparse')     |
| -        | 20.0±0.2ms           | 2.95±0.01ms         |    0.15 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(5, 'dense')      |
| -        | 23.1±0.3ms           | 622±2μs             |    0.03 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(5, 'sparse')     |
| -        | 480±2μs              | 134±0.8μs           |    0.28 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(1, 'sparse') |
| -        | 4.74±0.04ms          | 648±6μs             |    0.14 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(3, 'dense')  |
| -        | 5.23±0.03ms          | 137±0.7μs           |    0.03 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(3, 'sparse') |
| -        | 13.4±0.2ms           | 1.37±0.01ms         |    0.1  | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(5, 'dense')  |
| -        | 14.1±0.07ms          | 141±1μs             |    0.01 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(5, 'sparse') |
| -        | 133±0.6μs            | 76.6±0.4μs          |    0.57 | solve.TimeMatrixOperations.time_rref(3, 0)                           |
| -        | 248±0.3μs            | 89.2±0.3μs          |    0.36 | solve.TimeMatrixOperations.time_rref(4, 0)                           |
| -        | 24.1±0.1ms           | 10.2±0.04ms         |    0.42 | solve.TimeSolveLinSys189x49.time_solve_lin_sys                       |
| -        | 28.0±0.1ms           | 15.5±0.1ms          |    0.55 | solve.TimeSparseSystem.time_linsolve_Aaug(20)                        |
| -        | 54.3±0.4ms           | 25.0±0.04ms         |    0.46 | solve.TimeSparseSystem.time_linsolve_Aaug(30)                        |
| -        | 27.9±0.1ms           | 15.3±0.04ms         |    0.55 | solve.TimeSparseSystem.time_linsolve_Ab(20)                          |
| -        | 54.4±0.3ms           | 24.8±0.08ms         |    0.46 | solve.TimeSparseSystem.time_linsolve_Ab(30)                          |

Full benchmark results can be found as artifacts in GitHub Actions
(click on checks at the top of the PR).

oscarbenjamin · 2024-02-04T14:18:46Z

sympy/integrals/tests/test_meijerint.py

 def test_indefinite_1_bug():
    assert integrate((b + t)**(-a), t, meijerg=True
-        ) == -b**(1 - a)*(1 + t/b)**(1 - a)/(a - 1)
+        ).equals(-b**(1 - a)*(1 + t/b)**(1 - a)/(a - 1))


It would be better to change this test answer to show what is now returned.

The old result was:

In [4]: a, b, c = symbols("a b c") In [5]: integrate((b + t)**(-a), t, meijerg=True) Out[5]: 1 - a 1 - a ⎛ t⎞ -b ⋅⎜1 + ─⎟ ⎝ b⎠ ───────────────────── a - 1

The new result is

In [2]: integrate((b + t)**(-a), t, meijerg=True) Out[2]: 1 - a ⎛ t⎞ -b⋅⎜1 + ─⎟ ⎝ b⎠ ──────────────── a a a⋅b - b

oscarbenjamin · 2024-02-04T14:29:30Z

Looks like this also fixed gh-25949. A test could be added.

oscarbenjamin · 2024-02-04T14:33:46Z

I had wondered if this fixed gh-25137 but it does not seem to.

oscarbenjamin · 2024-02-05T12:32:09Z

This looks good. Thanks for this.

This is the kind of bug that could linger for a long time and I'm sure it took some work to find the proper fix.

25786 fix substitution

352587a

25786 add to mailmap

97890f0

oscarbenjamin added integrals integrals.meijerg labels Feb 4, 2024

25786 fix test

edc3e5b

maliesen marked this pull request as ready for review February 4, 2024 14:09

oscarbenjamin reviewed Feb 4, 2024

View reviewed changes

maliesen added 2 commits February 4, 2024 18:57

25786 make test assertion more explicit

fc9e939

25786 add test for issue sympy#25949

7f246d7

oscarbenjamin merged commit ee891ca into sympy:master Feb 5, 2024

oscarbenjamin mentioned this pull request Feb 5, 2024

Possible Bug in 'integrate' #25949

Closed

oscarbenjamin mentioned this pull request May 6, 2024

wrong result integrating sin(a*(x+pi))**2 #26566

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

[integrals] fix substitution formula in meijerint #26173

[integrals] fix substitution formula in meijerint #26173

Uh oh!

maliesen commented Feb 4, 2024

Uh oh!

sympy-bot commented Feb 4, 2024 •

edited

Loading

Uh oh!

oscarbenjamin commented Feb 4, 2024

Uh oh!

github-actions bot commented Feb 4, 2024 •

edited

Loading

Uh oh!

oscarbenjamin Feb 4, 2024

Uh oh!

oscarbenjamin commented Feb 4, 2024

Uh oh!

oscarbenjamin commented Feb 4, 2024

Uh oh!

oscarbenjamin commented Feb 5, 2024

Uh oh!

Uh oh!

Uh oh!

[integrals] fix substitution formula in meijerint #26173

[integrals] fix substitution formula in meijerint #26173

Uh oh!

Conversation

maliesen commented Feb 4, 2024

References to other Issues or PRs

Brief description of what is fixed or changed

Other comments

Release Notes

Uh oh!

sympy-bot commented Feb 4, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oscarbenjamin commented Feb 4, 2024

Uh oh!

github-actions bot commented Feb 4, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oscarbenjamin Feb 4, 2024

Choose a reason for hiding this comment

Uh oh!

oscarbenjamin commented Feb 4, 2024

Uh oh!

oscarbenjamin commented Feb 4, 2024

Uh oh!

oscarbenjamin commented Feb 5, 2024

Uh oh!

Uh oh!

sympy-bot commented Feb 4, 2024 •

edited

Loading

github-actions bot commented Feb 4, 2024 •

edited

Loading