GSOC 22' Project Discussion : Improving Series Expansions and Limit Computations

[Anutosh Bhat]

Hello All ,

This is Anutosh Bhat here . I'm currently pursuing an integrated dual degree in the field of Data Science from IIT Madras(2019 - 2024). I have been contributing to sympy since some time now and my work mainly revolves around the series, concrete, functions and  calculus module. A link to my work would be attached below.

During this journey of contributing to these modules I have a developed a good grasp of the current codebase. I also have a decent knowledge regarding what can be achieved currently through sympy and what are the areas where sympy could be improved.

Hence I have formulated a document (attached below) which includes all issues which could be tackled / new implementation / improvements we could be having in the series module over the summer . Some of these have solutions/relevant code prototypes/pseudocode attached or ordered steps needed to fix them . Other issues .... although I have mathematical clarity on them and have a decent idea of how they can be fixed still need careful observation from my side.  The ideas in the doc include

1. Issues caught by me and reported on the github page or issues caught while working on a Pr. ( A good chunk almost 70 % of this is doc is developed from my own experience , hence I have a good understanding of what's going wrong and what needs to be addressed)
2. Already existing issues on the github page/mailing list and issues /implementation reported by previous gsocer's in this module but weren't completed. ( 15% , I have somewhat clue of what needs to be fixed here and also their proposals give clear theory and implementation details to fix this)
3. New Implementation - These are things that have been pending since long in the series module for eg methods for ring series etc . This also includes methods influenced from other CAS which would be good additions to sympy. ( remaining 15%)

The purpose behind formulating this doc before building a proposal was that I have been thinking to apply for a 175 hour project (Short term variant) but I am not sure regarding some things which I am keen to have more clarity on . These are

1. What ideas out of this doc would sympy like to have and in which order ( ideas in the doc are not arranged according to priority)
2. Is there something else that  I have missed and should be addressed ??

Once I have clarity on this I'll start building a proposal . If this mail doesn't reach the potential mentor/co-mentors due to some reason I'll be contacting them through the email -id's provided in the idea list in the coming days .For now I'll like some guidance, reviews from not only the mentors but the whole community on the mailing list .

Thank you ,

Anutosh Bhat

My work - Pull requests · sympy/sympy (github.com)

Document - https://docs.google.com/document/d/1IG4tvXEnGuTRKzBLrSxdZf8TdtU13C_ZjSIDOdKW1vA/edit?usp=sharing

In Pdf format - 

[Anutosh Bhat]

  Hello All, 

This is Anutosh Bhat here . I have recently created a page for all my fellow contributors who plan to participate in Google Summer of Code 22' (GSoC 2022 Current Applications · sympy/sympy Wiki (github.com)) . I have also shared my draft proposal there . My proposal is not yet complete and I'll be working on it simultaneously as I get reviews .It misses key topics like Timeline and Milestones as of now which I would like to add with the help of the mentors .

The Proposal consists of 9 topics which are 

1. Improve leading term & series method for functions involving branch cuts/ branch points

2. Adding nseries, aseries and leading term support to special functions

3. Extending Functionality of Multivariate series expansions 

4. Ring series Improvements and Implementations

5. Extending coverage for limit_seq.py and formal.py

6. Fixing symbolic limit errors through gruntz and other gruntz errors

7. Refactor Order code and order.py to handle cases better

8. Improving AccumBounds for computing limits

9. Miscellaneous

These are all things and errors/areas which can be worked on currently in the series module to my knowledge and experience . I have discussed approaches , my ideas and pseudocode for a good chunk of these and will be developing these in the coming days but there are some things with which I am currently struggling with !!! and these are the somethings where I would like to have assistance from the mentors.

1. I am not sure which of these ideas topics would sympy like to have and the priority in which we should address these ideas , which is why I am struggling to decide which ideas could be kept for the future (future Gsoc periods and contributions in general) and which ones should be tackled in the upcoming Gsoc period. Hence I ended up explaining all the topics, the issues they have and how these can be approached to the fullest of my knowledge . Hence my proposed work is now ranging from page 12 to page 85.

2. Due to the first point I am further struggling to decide a time line to work as I can only propose the timeline once the priority of ideas has been decided.

3. As I would have a better idea regarding the time I would devote to each section once the priority has been confirmed.

4. Another factor hindering me from deciding a timeline is the project length. Although I would prefer a project length of 175 hours , I would be comfortable with any.

Hence I would request the mentors to read through the topics from page 12 .Before diving deep and discussing a topic.... a quick read through across all issues and topics would also be quite helpful as It would give you'll an idea regarding what would sympy prefer as of now (taking into consideration the current state of the series module) . I would be glad if the  mentors/co-mentors closely related to the series module and the org admins would help me out here .

Thanking You,

Yours Truly,

Anutosh Bhat.

IIT Madras(2019 - 2024).

[be19b037 Anutosh Surendra Bhat]

Hello All, 

This is Anutosh here again . I wanted to inform that  My Gsoc proposal (Improving series Expansions and Limit Computations) is almost about complete. I've received a basic review from the potential Mentor and also a decent to good review from the potential co-mentor .Hence, I would like to request for a second iteration of reviews from the org members , the community and members related to the series/limits module in particular . I would be quite confident in turning in my proposal on the Gsoc website , once this is done.

Obviously I'll be looking to reframe some things and look for any grammatical errors and mistakes but content wise I am pretty satisfied with what I have written down and don't see much change content wise . Obviously , if the are some changes requested I'll surely be addressing them too .

The proposal is exhaustive enough for a 175 hour project, to leave some room for bug fixes and implementations if I finish early . Hence I would also like to request if the members could validate the timeline proposed as that was a bit confusing for me to frame . Also now Gsoc projects can be extended to 22 weeks (midway) than the standard 12 week project , so like how do I consider or integrate this if required with the proposed timeline in my proposal?

All suggestions are welcome

Thanking You,

Yours Truly,

Anutosh Bhat

IIT MADRAS(2019 - 2024)

The link to the proposal is https://docs.google.com/document/d/1Cx7IaAkmxp1f81efSPNpMtkNE6K4NgkhkI_2TLjTEE8/edit#

In PDF format -

[Aaron Meurer]

On Sat, Apr 9, 2022 at 6:46 PM be19b037 Anutosh Surendra Bhat
<be19...@smail.iitm.ac.in> wrote:
>
> Hello All,
>

> This is Anutosh here again . I wanted to inform that My Gsoc proposal (Improving series Expansions and Limit Computations) is almost about complete. I've received a basic review from the potential Mentor and also a decent to good review from the potential co-mentor .Hence, I would like to request for a second iteration of reviews from the org members , the community and members related to the series/limits module in particular . I would be quite confident in turning in my proposal on the Gsoc website , once this is done.
>
> Obviously I'll be looking to reframe some things and look for any grammatical errors and mistakes but content wise I am pretty satisfied with what I have written down and don't see much change content wise . Obviously , if the are some changes requested I'll surely be addressing them too .
>
> The proposal is exhaustive enough for a 175 hour project, to leave some room for bug fixes and implementations if I finish early . Hence I would also like to request if the members could validate the timeline proposed as that was a bit confusing for me to frame . Also now Gsoc projects can be extended to 22 weeks (midway) than the standard 12 week project , so like how do I consider or integrate this if required with the proposed timeline in my proposal?

I'm not fully clear yet on how the new timelines work, but I would say
to just make the timeline in your proposal match the timeline you plan
to do for your project.

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 view this discussion on the web visit https://groups.google.com/d/msgid/sympy/31a576cc-823c-4000-b807-4a438d8fa68dn%40googlegroups.com.

[Anutosh Bhat]

Hello Aaron,

I hope you're doing well . I've tried to keep it simple and standard as how it has been since past couple years . But just for your information the Google open source blog says -  " You can spread the project out over a longer period of time and you can even switch to a longer timeframe mid-program if life happens. Rather than a mandatory 12-week program that runs from June – August with everyone required to finish their projects by the end of the 12th week, we are opening it up so mentors and their GSoC Contributors can decide together if they want to extend the deadline for the project up to 22 weeks " 

For me this seems to be useful if a selected contributor maybe falls short of time or maybe is stuck up with completion of some bits of the project or stretch goals or maybe wants to extend his project midway from a 175 hour project to a 350 hours project  but I'm not fully sure . Hence the question I had asked above ! 

Though ..... I've kept it simple as I plan to complete the project within the standard 12 week time span . Would also be glad if you could go through the proposal and let me know if there's something I could improve upon . Thanks for writing back anyways .I also submitted my proposal on the website too sometime back !

Thanking You,

Yours Truly,

Anutosh Bhat.

IIT MADRAS( 2019 - 2024)

[Aaron Meurer]

If I understand correctly then, the timeline is something that would
be decided on after you are selected (if you are) and assigned a
mentor, since it would also depend on the mentor's availability. So I
wouldn't worry about it for now, although it's definitely helpful to
note your plans in your proposal so that we can be aware of them.

Aaron Meurer

> To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/c380d13d-2b7f-4130-b0f3-0ba79ae00fe0n%40googlegroups.com.