Skip to content

perf(matrices): make values etc faster for sparse matrices (1.13) #26674

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 6 commits into from
Jun 5, 2024
Merged

perf(matrices): make values etc faster for sparse matrices (1.13) #26674

merged 6 commits into from
Jun 5, 2024

Conversation

oscarbenjamin
Copy link
Collaborator

References to other Issues or PRs

Backport of gh-26058 to 1.13 release branch.

Brief description of what is fixed or changed

Other comments

Release Notes

  • matrices
    • Two new Matrix methods iter_values and iter_items are added for lazy iteration over the nonzero elements of a Matrix. A Matrix method from_dok is added for efficient direct construction of sparse matrices. Some basic matrix operations are made faster by using these methods.

@sympy-bot
Copy link

sympy-bot commented Jun 5, 2024 •
edited
Loading

✅

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:

  • matrices
    • Two new Matrix methods iter_values and iter_items are added for lazy iteration over the nonzero elements of a Matrix. A Matrix method from_dok is added for efficient direct construction of sparse matrices. Some basic matrix operations are made faster by using these methods. (#26674 by @oscarbenjamin)

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. -->

Backport of gh-26058 to 1.13 release branch.

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


#### 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:

* solvers
  * Added a new solver for logarithmic equations.

* functions
  * Fixed a bug with log of integers. Formerly, `log(-x)` 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 -->
* matrices
   * Two new Matrix methods `iter_values` and `iter_items` are added for lazy iteration over the nonzero elements of a `Matrix`. A Matrix method `from_dok` is added for efficient direct construction of sparse matrices. Some basic matrix operations are made faster by using these methods.
<!-- END RELEASE NOTES -->

Update

The release notes on the wiki have been updated.

@oscarbenjamin oscarbenjamin added this to the SymPy 1.13 milestone Jun 5, 2024
@github-actions GitHub Actions
Copy link

github-actions bot commented Jun 5, 2024

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 [a36a8b23] <sympy-1.12.1^0>   | After [7ba1ae46]    |   Ratio | Benchmark (Parameter)                                                |
|----------|--------------------------------------|---------------------|---------|----------------------------------------------------------------------|
| -        | 69.4±0.6ms                           | 44.5±0.3ms          |    0.64 | integrate.TimeIntegrationRisch02.time_doit(10)                       |
| -        | 67.8±1ms                             | 44.0±0.5ms          |    0.65 | integrate.TimeIntegrationRisch02.time_doit_risch(10)                 |
| +        | 19.0±0.1μs                           | 30.2±0.1μs          |    1.59 | integrate.TimeIntegrationRisch03.time_doit(1)                        |
| -        | 73.4±1ms                             | 29.0±0.4ms          |    0.4  | polys.TimeGCD_GaussInt.time_op(1, 'dense')                           |
| -        | 26.0±0.2ms                           | 17.0±0.06ms         |    0.66 | polys.TimeGCD_GaussInt.time_op(1, 'expr')                            |
| -        | 75.0±0.3ms                           | 29.1±0.2ms          |    0.39 | polys.TimeGCD_GaussInt.time_op(1, 'sparse')                          |
| -        | 258±2ms                              | 128±1ms             |    0.5  | polys.TimeGCD_GaussInt.time_op(2, 'dense')                           |
| -        | 258±2ms                              | 127±2ms             |    0.49 | polys.TimeGCD_GaussInt.time_op(2, 'sparse')                          |
| -        | 654±5ms                              | 377±2ms             |    0.58 | polys.TimeGCD_GaussInt.time_op(3, 'dense')                           |
| -        | 663±5ms                              | 381±3ms             |    0.58 | polys.TimeGCD_GaussInt.time_op(3, 'sparse')                          |
| -        | 501±4μs                              | 291±1μs             |    0.58 | polys.TimeGCD_LinearDenseQuadraticGCD.time_op(1, 'dense')            |
| -        | 1.79±0.01ms                          | 1.03±0.01ms         |    0.57 | polys.TimeGCD_LinearDenseQuadraticGCD.time_op(2, 'dense')            |
| -        | 5.83±0.07ms                          | 3.07±0.02ms         |    0.53 | polys.TimeGCD_LinearDenseQuadraticGCD.time_op(3, 'dense')            |
| -        | 446±1μs                              | 230±1μs             |    0.52 | polys.TimeGCD_QuadraticNonMonicGCD.time_op(1, 'dense')               |
| -        | 1.46±0.01ms                          | 668±3μs             |    0.46 | polys.TimeGCD_QuadraticNonMonicGCD.time_op(2, 'dense')               |
| -        | 4.93±0.05ms                          | 1.67±0.01ms         |    0.34 | polys.TimeGCD_QuadraticNonMonicGCD.time_op(3, 'dense')               |
| -        | 378±2μs                              | 205±0.4μs           |    0.54 | polys.TimeGCD_SparseGCDHighDegree.time_op(1, 'dense')                |
| -        | 2.44±0.01ms                          | 1.23±0ms            |    0.5  | polys.TimeGCD_SparseGCDHighDegree.time_op(3, 'dense')                |
| -        | 10.1±0.05ms                          | 4.37±0.03ms         |    0.43 | polys.TimeGCD_SparseGCDHighDegree.time_op(5, 'dense')                |
| -        | 358±1μs                              | 169±2μs             |    0.47 | polys.TimeGCD_SparseNonMonicQuadratic.time_op(1, 'dense')            |
| -        | 2.49±0.01ms                          | 899±3μs             |    0.36 | polys.TimeGCD_SparseNonMonicQuadratic.time_op(3, 'dense')            |
| -        | 9.72±0.06ms                          | 2.71±0.01ms         |    0.28 | polys.TimeGCD_SparseNonMonicQuadratic.time_op(5, 'dense')            |
| -        | 1.04±0.01ms                          | 428±2μs             |    0.41 | polys.TimePREM_LinearDenseQuadraticGCD.time_op(3, 'dense')           |
| -        | 1.72±0.01ms                          | 505±2μs             |    0.29 | polys.TimePREM_LinearDenseQuadraticGCD.time_op(3, 'sparse')          |
| -        | 5.98±0.08ms                          | 1.81±0.02ms         |    0.3  | polys.TimePREM_LinearDenseQuadraticGCD.time_op(5, 'dense')           |
| -        | 8.52±0.03ms                          | 1.49±0.01ms         |    0.18 | polys.TimePREM_LinearDenseQuadraticGCD.time_op(5, 'sparse')          |
| -        | 284±2μs                              | 65.4±0.2μs          |    0.23 | polys.TimePREM_QuadraticNonMonicGCD.time_op(1, 'sparse')             |
| -        | 3.47±0.04ms                          | 390±2μs             |    0.11 | polys.TimePREM_QuadraticNonMonicGCD.time_op(3, 'dense')              |
| -        | 3.99±0.05ms                          | 281±0.6μs           |    0.07 | polys.TimePREM_QuadraticNonMonicGCD.time_op(3, 'sparse')             |
| -        | 6.98±0.01ms                          | 1.25±0.01ms         |    0.18 | polys.TimePREM_QuadraticNonMonicGCD.time_op(5, 'dense')              |
| -        | 8.70±0.07ms                          | 858±8μs             |    0.1  | polys.TimePREM_QuadraticNonMonicGCD.time_op(5, 'sparse')             |
| -        | 4.96±0.01ms                          | 2.98±0ms            |    0.6  | polys.TimeSUBRESULTANTS_LinearDenseQuadraticGCD.time_op(2, 'sparse') |
| -        | 12.2±0.1ms                           | 6.69±0.04ms         |    0.55 | polys.TimeSUBRESULTANTS_LinearDenseQuadraticGCD.time_op(3, 'dense')  |
| -        | 22.3±0.1ms                           | 9.07±0.02ms         |    0.41 | polys.TimeSUBRESULTANTS_LinearDenseQuadraticGCD.time_op(3, 'sparse') |
| -        | 5.20±0.01ms                          | 868±7μs             |    0.17 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(1, 'sparse')    |
| -        | 12.5±0.05ms                          | 7.03±0.06ms         |    0.56 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(2, 'sparse')    |
| -        | 103±1ms                              | 25.7±0.09ms         |    0.25 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(3, 'dense')     |
| -        | 166±3ms                              | 53.9±0.07ms         |    0.33 | polys.TimeSUBRESULTANTS_QuadraticNonMonicGCD.time_op(3, 'sparse')    |
| -        | 174±0.8μs                            | 112±0.9μs           |    0.64 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(1, 'dense')      |
| -        | 361±4μs                              | 215±1μs             |    0.6  | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(1, 'sparse')     |
| -        | 4.24±0.03ms                          | 857±3μs             |    0.2  | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(3, 'dense')      |
| -        | 5.26±0.05ms                          | 387±1μs             |    0.07 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(3, 'sparse')     |
| -        | 20.8±0.2ms                           | 2.82±0.01ms         |    0.14 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(5, 'dense')      |
| -        | 22.9±0.2ms                           | 636±5μs             |    0.03 | polys.TimeSUBRESULTANTS_SparseGCDHighDegree.time_op(5, 'sparse')     |
| -        | 482±0.9μs                            | 134±0.6μs           |    0.28 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(1, 'sparse') |
| -        | 4.73±0.08ms                          | 621±4μs             |    0.13 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(3, 'dense')  |
| -        | 5.23±0.04ms                          | 139±2μs             |    0.03 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(3, 'sparse') |
| -        | 13.0±0.1ms                           | 1.31±0.01ms         |    0.1  | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(5, 'dense')  |
| -        | 13.9±0.3ms                           | 141±1μs             |    0.01 | polys.TimeSUBRESULTANTS_SparseNonMonicQuadratic.time_op(5, 'sparse') |
| -        | 132±0.9μs                            | 75.1±0.3μs          |    0.57 | solve.TimeMatrixOperations.time_rref(3, 0)                           |
| -        | 251±2μs                              | 89.4±0.1μs          |    0.36 | solve.TimeMatrixOperations.time_rref(4, 0)                           |
| -        | 24.3±0.1ms                           | 10.1±0.03ms         |    0.42 | solve.TimeSolveLinSys189x49.time_solve_lin_sys                       |
| -        | 28.8±0.2ms                           | 15.6±0.08ms         |    0.54 | solve.TimeSparseSystem.time_linsolve_Aaug(20)                        |
| -        | 55.1±0.4ms                           | 24.8±0.05ms         |    0.45 | solve.TimeSparseSystem.time_linsolve_Aaug(30)                        |
| -        | 28.4±0.07ms                          | 15.3±0.1ms          |    0.54 | solve.TimeSparseSystem.time_linsolve_Ab(20)                          |
| -        | 55.2±0.3ms                           | 24.6±0.1ms          |    0.45 | 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
  oscarbenjamin



        merged commit 8c2d903
      into

      
  
    sympy:1.13
  


    Jun 5, 2024

  







  

  
  

  
      @oscarbenjamin
  oscarbenjamin


    
    deleted the
    
      
        pr_matrix_values_113
    
    branch

    June 5, 2024 13:04    
    













  
  

    

      

  




      


    


    

      

        
      

  
    Sign up for free
    to join this conversation on GitHub.
    Already have an account?
    Sign in to comment


  



      

    

  




  
          



      

  

    
    

    

  

    Reviewers
  


    

    No reviews






    

  


      
  

    Assignees
  



        


    No one assigned







      


  


  

    Labels
  



    

      

    domainmatrix

  

    matrices

  

    Performance









      

  

    

        

    Projects
  


        




    None yet




  



      


  

    
  

    Milestone
  


      
    
  
    SymPy 1.13
  





    
      
          


  

    

      
        

          
  

    Development
  



            


Successfully merging this pull request may close these issues.





  
  



      
    

  




      

    

  
      

  

    

      2 participants
    

    

        
          @oscarbenjamin 
        
          @sympy-bot