Exact (LISP-ish) calculations in Ruby?

A

Aldric Giacomoni

Has anyone written a gem for exact calculations? The kind one would find
on a LISP REPL, or when using your average HP scientific calculator?
I couldn't find anything on gemcutter or rubyforge, but maybe I missed
it..
I don't ask because I need it.. I ask because of an idealistic desire
for exact calculations :)
 
M

Marnen Laibow-Koser

Aldric said:
Has anyone written a gem for exact calculations? The kind one would find
on a LISP REPL, or when using your average HP scientific calculator?

I couldn't find anything on gemcutter or rubyforge, but maybe I missed
it..
I don't ask because I need it.. I ask because of an idealistic desire
for exact calculations :)

Wiring Rational into Numeric#/ shouldn't be too hard. I'm not sure how
to deal with square roots without lots of special cases, however.


Best,
 
D

David Masover

Has anyone written a gem for exact calculations? The kind one would find
on a LISP REPL, or when using your average HP scientific calculator?


=> 3/4

This much is available through the Rational object:

irb(main):001:0> Rational 3,4
=> (3/4)
irb(main):002:0> Rational(3) / 4
=> (3/4)
irb(main):003:0> Rational(3) / 4 / 12
=> (1/16)

I'm not sure about the rest of it, though it makes me wish I'd used Lisp for a
recent project. I pretty much had to invent a symbolic equation manipulation
class -- though, to be fair, I didn't look that hard for an existing one.
 
A

Aldric Giacomoni

David said:
irb(main):002:0> Rational(3) / 4
=> (3/4)
irb(main):003:0> Rational(3) / 4 / 12
=> (1/16)

I'm not sure about the rest of it, though it makes me wish I'd used Lisp
for a recent project.

In all fairness, I just fired up SLIME to check, and LISP's (sqrt 2)
does come out to be an approximation, so I guess I want something cool
like the HP-48 and HP-49's factoring power..
 
A

Aldric Giacomoni

Aldric said:
In all fairness, I just fired up SLIME to check, and LISP's (sqrt 2)
does come out to be an approximation, so I guess I want something cool
like the HP-48 and HP-49's factoring power..

And I just read up on the process of 'continuing fractions' which can be
used find fractions / estimates of irrational numbers down to the nth
decimal. I wonder how we can know that, for instance, 1.41421 (etc) is
sqrt(2) ... ?

Ah, mathematics.
 
K

Ken Bloom

Wiring Rational into Numeric#/ shouldn't be too hard. I'm not sure how
to deal with square roots without lots of special cases, however.

Umm, there's a standard library for wiring in Rational:

require 'mathn'
 
G

Gavin Sinclair

Umm, there's a standard library for wiring in Rational:

require 'mathn'

You've gotta love this...

$ irb=> 34

(It's just a printing "error".)
=> 0.75

That's Ruby 1.8.7.
 
G

Gavin Sinclair

And I just read up on the process of 'continuing fractions' which can be
used find fractions / estimates of irrational numbers down to the nth
decimal. I wonder how we can know that, for instance, 1.41421 (etc) is
sqrt(2) ... ?

There's an algorithm for calculating square roots, similar (in
appearance, not in process) to the algorithm for long division. One
or more generations ago, school students would have done it with pen
and paper, I think.

Furthermore, there's a Calculus-based algorithm (Newton's method, it's
called in my syllabus, but I think it's properly called the Newton-
Raphson method) for calculating square/cube/fourth/... roots to any
desired accuracy.

Thirdly, you can estimate any root you like by trial multiplication!
(Exercise: write a Ruby program to do this; shouldn't take more than
10 lines.)
 
G

Gavin Sinclair

Has anyone written a gem for exact calculations? The kind one would find
on a LISP REPL, or when using your average HP scientific calculator?


I couldn't find anything on gemcutter or rubyforge, but maybe I missed
it..
I don't ask because I need it.. I ask because of an idealistic desire
for exact calculations :)

That "idealistic desire" would only be needed in very special cases in
real programming, so it's not surprising not to find a Ruby library
for it.

Essentially, you're looking for the capabilities provided by a
Computer Algebra System (CAS) like Mathematica, as mentioned, or
Maxima (open source, IIRC). You may be able to find an open source
CAS that has Ruby bindings, or for which (if you're sufficiently
motivated) you can write some bindings.

Or you can give it a go!
sqrt(2) * sqrt(3)
--> sqrt(6)
--> 2 * sqrt(2)
2 * PI * 3.5
--> 7 * PI

Not my idea of fun, though. The beauty of Mathematics is best
appreciated in the mind, for it's only there that sqrt(2) exists.
 
M

Martin DeMello

There's an algorithm for calculating square roots, similar (in
appearance, not in process) to the algorithm for long division. =A0One
or more generations ago, school students would have done it with pen
and paper, I think.

I learnt it in school, and I'm in my 30s, so one generation ago.

martin
 
F

Florian Frank

Aldric said:
And I just read up on the process of 'continuing fractions' which can be
used find fractions / estimates of irrational numbers down to the nth
decimal. I wonder how we can know that, for instance, 1.41421 (etc) is
sqrt(2) ... ?

Ah, mathematics.
I had to implement them for my bullshit library (which has grown to
include half a math library by now, ugh):

http://github.com/flori/bullshit/blob/master/lib/bullshit.rb#L20

require 'bullshit' # pardon my french
include Bullshit
ContinuedFraction.for_a { |n| n == 0 ? 1 : 2 }[]
# => 1.4142135623731

They are rather fascinating. They can be used to implement functions as
well, here's the atan function:

atan = ContinuedFraction.for_a do |n, x|
n == 0 ? 0 : 2 * n - 1
end.for_b do |n, x|
n <= 1 ? x : ((n - 1) * x) ** 2
end

# => 0.785398163397448

Now, it's easy to approximate pi as well:

pi = lambda { 4 * atan[1] }
# => 3.14159265358979

As you can imagine I had a lot of fun playing with them. ;)
 
A

Aldric Giacomoni

Gavin said:
Not my idea of fun, though. The beauty of Mathematics is best
appreciated in the mind, for it's only there that sqrt(2) exists.

Gavin, I'll take a look at existing CAS :) This being said, I rather
think that the ability to use mental concepts without worrying that the
computer will change them is rather important (I -said- sqrt(2), not
1.414 !) .. And, well, if Ruby can do it, then it should be available
for those who do need it, don't you think? :)

Florian said:
I had to implement them for my bullshit library (which has grown to
include half a math library by now, ugh):
http://github.com/flori/bullshit/blob/master/lib/bullshit.rb#L20

Wow, massive amount of work you did there. Pretty cool coding, too. I
may just have to shamelessly steal some of these :)
 
M

Marnen Laibow-Koser

Martin said:
I learnt it in school, and I'm in my 30s, so one generation ago.

Yeah, I learned it as a kid as well, and I'm about the same age. I've
long since forgotten how to do it, though; I do remember that it
involves a bit of trial multiplication.


Best,
 
F

Florian Frank

Aldric said:
Wow, massive amount of work you did there. Pretty cool coding, too. I
may just have to shamelessly steal some of these :)
Oh, thanks. I have considered pulling some of the math stuff out of the
library and putting it in an extra gem. There are some algorithms in
there, that could benefit from being being transformed into a c
extension as well. On the other hand I was pretty surprised how well
Ruby did on the numerical stuff and also how much better Ruby 1.9 was on
the algorithms that iterate a lot.
 
D

David Masover

Furthermore, there's a Calculus-based algorithm (Newton's method, it's
called in my syllabus, but I think it's properly called the Newton-
Raphson method) for calculating square/cube/fourth/... roots to any
desired accuracy.

Newton's Method, according to at least one Calculus textbook, is a way of
finding roots (zeros) for any function for which you can calculate function
values and derivatives. The example given is taking the cube root of seven, by
rewriting the problem as:

x^3 - 7 = 0

The derivative of which is easy to calculate as 3x^2.

I wrote a program to do Newton's Method. It's one of the few times I wished I
was using Lisp instead of Ruby, as I had no easy way of taking apart a block
as source code to find its derivative. Instead, whenever I apply it, I have to
take the derivative manually, or feed it through Maxima.

But I guess if what you wanted was something that knows how to do algebraic
manipulation, without losing accuracy until you tell it to give you a float,
there's always Maxima.
 
G

Gavin Sinclair

Newton's Method, according to at least one Calculus textbook, is a way of
finding roots (zeros)

_estimating_ roots (to any desired accuracy)
for any function for which you can calculate function
values and derivatives.
I wrote a program to do Newton's Method. It's one of the few times I wished I
was using Lisp instead of Ruby, as I had no easy way of taking apart a block
as source code to find its derivative. Instead, whenever I apply it, I have to
take the derivative manually, or feed it through Maxima.

Interesting. I'd be curious to see Lisp and Ruby approaches to
representing and differentiating functions. Polynomials would be
trivial in Ruby if you build an appropriate representation, but
general functions? Leave that to the experts, I guess.
 
D

David Masover

_estimating_ roots (to any desired accuracy)

Good point -- though, technically, unless it fails utterly (which it sometimes
does), you can find the _exact_ root, given infinite time to calculate it :p
I'd be curious to see Lisp and Ruby approaches to
representing and differentiating functions. Polynomials would be
trivial in Ruby if you build an appropriate representation, but
general functions? Leave that to the experts, I guess.

I don't think I'd have problems differentiating most _mathematical_
expressions, given an appropriate description. I could pretty much just copy
techniques out of the book -- chain rule, product rule, etc. That said, math
has been around for thousands of years, and I'm sure there's something I
haven't thought of, or even learned yet.

But yes, polynomials are definitely where it's easy -- I wrote something to
integrate arbitrary polynomials. In order to do this, I had to write a fairly
messy library for handling mathematical expressions. I'm thinking I could
probably go back over it and clean up the object model, at least.

The main reason this would be easier in Lisp is that rather than building said
messy object model to hold the expressions, I'd just work with sexps. Macros
would let me actually do something like this:

(differentiate (- (expt x 3) 7))

If we could do the equivalent in Ruby, it'd look like this:

Math.differentiate {|x| x**3 - 7}

That's kind of gross to implement, though. Pure does it by discarding the
block, finding the original source, and re-parsing it, using whatever parse
tool is available, to get the actual parse tree.

I decided to start with the object model, and try to add a DSL later.
The following actually works:

irb(main):004:0> (E:)x)**3 - 7).integrate :x
=> ((x^4/4)+((-7)*x))

Again, only with polynomials. The first big challenge here was to get it into a
reasonable representation. The second was working with that representation,
while retaining some sanity.

But you can see where I'm blatantly cheating above. It could be made easier to
work with by messing with Symbol, but it still has the fatal flaw that I'm
faking it -- you're still not actually typing in an expression, you're typing
a recipe for constructing an Expression object tree, whereas in Lisp, your
sexp would already be the tree I'm looking for.

And the motivation? Numeric integration using a lagrange polynomial. Only for
fun -- this is too obvious an idea not to have been tried already. Once I have
my algebraic-manipulation-and-integration library, it's about 30 lines of code
to integrate an arbitrary set of points with this method.

If anyone actually wants this code, I'll throw it on github. The main reason I
haven't is that I've got to be reinventing like five wheels here. Plus, it
could use some cleaning up -- I haven't really made an effort to unify my
Function library (which does things like Newton's method, Simpson's rule, etc)
with my Expression library (which does the above hackery).

But really, I've done this as a learning exercise. As a tool, I'd still
probably use Maxima.
 
A

Aldric Giacomoni

David said:
Macros would let me actually do something like this:

(differentiate (- (expt x 3) 7))

If we could do the equivalent in Ruby, it'd look like this:

Math.differentiate {|x| x**3 - 7}

That's kind of gross to implement, though. Pure does it by discarding
the
block, finding the original source, and re-parsing it, using whatever
parse
tool is available, to get the actual parse tree.

I decided to start with the object model, and try to add a DSL later.
The following actually works:

irb(main):004:0> (E:)x)**3 - 7).integrate :x
=> ((x^4/4)+((-7)*x))
[...]
But really, I've done this as a learning exercise. As a tool, I'd still
probably use Maxima.

Everyone's talking about Maxima, which was written in LISP.. Interesting
:)
I'd like to modify the earlier question then, and wonder if Ruby is the
right tool for the job, or if one should simply just go with LISP (in
which case, there's no reinventing the wheel, I'd just use Maxima). If
Ruby can be the right tool for the job.. Is it worth it? :)

I do think it would be pretty cool to have a CAS like Maxima in Ruby
(there is in fact a rubyforge project to develop one, but it hasn't
released any files at all yet).
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,997
Messages
2,570,239
Members
46,827
Latest member
DMUK_Beginner

Latest Threads

Top