info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Rubi 5 Development Prototype Available
59 views
Skip to first unread message
Albert Rich
Mar 18, 2021, 4:42:18 AM3/18/21
to
A newly created repository on GitHub for Rubi 5 at
https://github.com/RuleBasedIntegration/Rubi-5
contains an actual functioning prototype of Rubi 5 that shows the structure of the proposed integrator. Also it provides an example how Rubi 4 pattern matching rules can be compiled into a Rubi 5 rule-based if-then-else decision tree. The repository's README file provides more details.
https://github.com/RuleBasedIntegration/Rubi-5
contains an actual functioning prototype of Rubi 5 that shows the structure of the proposed integrator. Also it provides an example how Rubi 4 pattern matching rules can be compiled into a Rubi 5 rule-based if-then-else decision tree. The repository's README file provides more details.
Message has been deleted
Albert Rich
Mar 19, 2021, 9:33:03 PM3/19/21
to
A question was asked on Rubi's Gitter chatroom (https://gitter.im/Rule-Based-Integration/Lobby) about the difficulty of implementing the utility routines Rubi 5 uses on a host CAS; in particular the Simplify function. I posted the following response:
The Simplify function of the host CAS is not critical to Rubi 5 producing mathematically correct antiderivatives. It only affects how neat and tidy they are.
The Simplify function of the host CAS is not critical to Rubi 5 producing mathematically correct antiderivatives. It only affects how neat and tidy they are.
sciw...@gmail.com
May 14, 2021, 1:11:00 AM5/14/21
to
On Saturday, March 20, 2021 at 2:33:03 PM UTC+13, Albert Rich wrote:
> A question was asked on Rubi's Gitter chatroom (https://gitter.im/Rule-Based-Integration/Lobby) about the difficulty of implementing the utility routines Rubi 5 uses on a host CAS; in particular the Simplify function. I posted the following response:
>
> The Simplify function of the host CAS is not critical to Rubi 5 producing mathematically correct antiderivatives. It only affects how neat and tidy they are.
This is quite fascinating, in a previous post you mentioned a compiler for RUBI, what type of compiler might this be; I've heard
> A question was asked on Rubi's Gitter chatroom (https://gitter.im/Rule-Based-Integration/Lobby) about the difficulty of implementing the utility routines Rubi 5 uses on a host CAS; in particular the Simplify function. I posted the following response:
>
> The Simplify function of the host CAS is not critical to Rubi 5 producing mathematically correct antiderivatives. It only affects how neat and tidy they are.
of the Fortran compiler.
Albert Rich
May 14, 2021, 9:22:10 PM5/14/21
to
On Thursday, May 13, 2021 at 7:11:00 PM UTC-10, sciw...@gmail.com wrote:
> This is quite fascinating, in a previous post you mentioned a compiler for RUBI, what type of
> compiler might this be; I've heard of the Fortran compiler.
pdf files on the GitHub repository for Rubi 5 at https://github.com/RuleBasedIntegration/Rubi-5
> This is quite fascinating, in a previous post you mentioned a compiler for RUBI, what type of
> compiler might this be; I've heard of the Fortran compiler.
show the result of compiling Rubi 4 pattern matching rules into Rubi 5 functions defined using a single, deeply nested if-then-else control construct.
The files "1.1.1 (a+b x)^m.pdf" and "1.2.1 (a+b x+c x^2)^p.pdf" show the numerous Rubi 4 pattern matching rules required to integrate expressions of the form (a+b x)^m and (a+b x+c x^2)^p, respectively.
The files "Rule 1.1.1 Int[(a+b x)^m,x].pdf" and "Rule 1.2.1 Int[(a+b x+c x^2)^p,x].pdf" show the equivalent Rubi 5 functions Int111 and Int121, each defined by a single if-then-else control construct.
Comparing the Rubi 4 and 5 pdf files side-by-side should clarify what compilation entails.
Albert
0 new messages