GSOC 2016 : Risch algorithm for symbolic integration

[rishabh daal]

Hi,

I am Rishabh Daal , 3rd year Undergraduate in Mathematics and Computing from IIT Varanasi. I wish to contribute to SymPy and i am interested in Risch algorithm for symbolic integration.

I have gone through   "Symbolic Integration Tutorial" and first chapter of Bronstein's book and everything looks interesting.

Though i have gone through  last active thread, it will be great if someone can give me a brief account on

1. Current status of the project and people working on it .

2. What has been done till now  ?

3. What is expected to be done ?

[Aaron Meurer]

Nothing has really happened since this thread. No one is working on it.

Aaron Meurer

On Sun, Mar 6, 2016 at 5:57 AM, rishabh daal <risha...@gmail.com> wrote:
> HI,

> I am Rishabh Daal , 3rd year Undergraduate in Mathematics and Computing from
> IIT Varanasi. I wish to contribute to SymPy and i am interested in Risch
> algorithm for symbolic integration.
> I have gone through "Symbolic Integration Tutorial" and first chapter of

> Bronstein's book and everything looks fascinating.
>
> Though i have gone through last active thread , it will be great if someone

> give me a brief account on
>

> 1. what is Current status of the project and people working on it .
>
> 2. What has been done after ?

>
> 3. What is expected to be done ?
>

> --
> You received this message because you are subscribed to the Google Groups
> "sympy" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sympy+un...@googlegroups.com.
> To post to this group, send email to sy...@googlegroups.com.
> Visit this group at https://groups.google.com/group/sympy.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sympy/c4413878-8bf4-4190-aff5-fb75cdabe20b%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Abhishek Verma]

 ohkk so no one working on this idea right now. but as Mr.  Rishabh Daal said that he wants to work on this.And I also wants work on this idea . So Actually I want to know the what's the plan of  sympy community for Risch algorithm for symbolic integration for Gsoc 2016.I would also wants to know
 Exactly what is present state of this of idea.Hope Rishabh Daal also wants answer of my question.
Cheers
Abhishek Verma

[Aaron Meurer]

The goal for the Risch algorithm is to complete the implementation.
Finishing Bronstein's book (i.e., integration of trigonometric
functions) ought to be enough to fill a summer. Beyond that one could
also implement support for some special functions, such as erf.

One could also work on implementing the algebraic case of the
algorithm. This is more work, and requires more mathematical
background.

The current status is, sadly, exactly the same as it was when that
thread that Rashabh linked to was written.

Aaron Meurer

> Cheers
> Abhishek Verma

>
> --
> You received this message because you are subscribed to the Google Groups
> "sympy" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sympy+un...@googlegroups.com.
> To post to this group, send email to sy...@googlegroups.com.
> Visit this group at https://groups.google.com/group/sympy.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sympy/07f68e5e-0f36-4cea-b755-9506525d252e%40googlegroups.com.

[Abhishek Verma]

Ohkkk Right now I am studying Bronstein's book. And also making abstract algebra part stronger. I also want to Implement mixed algebraic-
transcendental functions integration apart from algebraic and trigonometric function. 

Sir Aaron Meurer , As you say that its status is bad, so if I am going to apply for gsoc and making my application form ,should I focus on Completion Risch Algorithm Implementation or trying enchancing the Integration part??

I know It  will require very very Strong Mathematical Background, so actually I am trying to asking that how exactly we should prepare Ourself to applying for Gsoc 2016?
How we can make our goal very practical in our application form??

Thanks in Advance Sir
Cheers
Abhishek Verma

[Aaron Meurer]

On Tuesday, March 8, 2016, Abhishek Verma <iamverma...@gmail.com> wrote:

Ohkkk Right now I am studying Bronstein's book. And also making abstract algebra part stronger. I also want to Implement mixed algebraic-
transcendental functions integration apart from algebraic and trigonometric function. 

Sir Aaron Meurer , As you say that its status is bad, so if I am going to apply for gsoc and making my application form ,should I focus on Completion Risch Algorithm Implementation or trying enchancing the Integration part??

It's not necessarily bad. The algorithm does work, and is used in SymPy. But only for the class of functions that have been implemented (a subset of the exponential and logarithmic cases). The bad thing is that little to no work has been done on the algorithm for a while. 

 

I know It  will require very very Strong Mathematical Background, so actually I am trying to asking that how exactly we should prepare Ourself to applying for Gsoc 2016?
How we can make our goal very practical in our application form??

Read through Bronstein's book and make sure you can understand the material. Also read through the code. Most of it is based on the pseudo code from the book but not all of it. 

A good place to start would be to try to finish these unfinished pull requests https://github.com/sympy/sympy/pulls?q=is%3Apr+is%3Aopen+risch+author%3Acheatiiit.

Aaron Meurer

Thanks in Advance Sir
Cheers
Abhishek Verma

 

 

--

You received this message because you are subscribed to the Google Groups "sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sympy+un...@googlegroups.com.
To post to this group, send email to sy...@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.

To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/b4eb089f-0e11-4c26-b441-8e7da5c05950%40googlegroups.com.

[Abhishek Verma]

Sir  I have also posted an issuse on github https://github.com/sympy/sympy/issues/10680 I don't know ,what's the importance of that but I want to fix that , can you see that ,if you have to suugest something ,I would welcome your anyone suggestion happily  :))

and If you have something to suggest us, that we shall required for deep understanding of Bronstien Books Topic then you can suggest.
We required better and effective resources for our study. 

[Aaron Meurer]

Just to be clear, that issue has to do with the heurisch algorithm,
which is not the same as the risch algorithm. The heurisch algorithm
is a heuristic version of the Risch algorithm whose implementation
predates the Risch algorithm's in SymPy. If you search the issue
tracker for "risch" you can find some issues in the Risch algorithm.
You can also look for TODOs in the code. Also, if it's your first
contribution, it might be worth starting with an "Easy to Fix" issue,
even if it's unrelated to integration.

Aaron Meurer

> --
> You received this message because you are subscribed to the Google Groups
> "sympy" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sympy+un...@googlegroups.com.
> To post to this group, send email to sy...@googlegroups.com.
> Visit this group at https://groups.google.com/group/sympy.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sympy/f3e9f2b5-a126-49b7-a841-3a2524d530c7%40googlegroups.com.