Sirb¶ ↑

• github.com/davidrichards/sirb/tree/master

Statistics + IRB. This offers a series of useful tools that a console should probably have, if your goal is to crunch a few numbers. It includes all the packages that I use, if you have them. Statisticus will have a standard library of statistical methods that you may want to have access to, so it’s probably worth loading that gem as well.

Sirb has a set of rather useful functions setup:

• max

• min

• sum

• mean

• median

• range

• variance (also var)

• standard_deviation (also std)

• sort

• rank

• order

• quantile

• cumulative_sum (also cum_sum)

• cumulative_product (also cum_prod)

• cumulative_max (also cum_max)

• cumulative_min (also cum_min)

For comparing lists, you can use:

• correlation (also cor)

• p_max

• p_min

For other things, you can use:

• max

• min

• product

• to_pairs

Right now, Sirb attempts to load the following libraries (when available):

• NArray

• RGL

• matrix

• set

• RNum

• Statisticus

• RBTree

• mathn

• This library

As you work, you can choose to store a procedure and use it between sessions. I do this with pstore (part of Ruby’s standard library) and some code I found on Ruby Quiz by Florian Groß. The storage process is a little draconian, so in the short-term, you can call sirb with -w (sirb -w) and this feature is taken away from the runtime.

This is a very useful tool to have if you have some R code that you tend to use often, say, or some other functions that you discover while working with some data.

From command line:

sirb
>> @a = [3,2,6,7,14]
=> [3, 2, 6, 7, 14]
>> @b = [4,6,2,1,19]
=> [4, 6, 2, 1, 19]
>> @a.max 
=> 14
>> @a.min 
=> 2
>> @a.sum
=> 32.0
>> # Most methods have meaningful block semantics
?> @a.sum {|x| x ** x}
=> 1.11120068264282e+16
>> @a.mean
=> 6.4
>> @a.median 
=> 6
>> @a.range  
=> [2, 14]
>> @a.var
=> 22.3
>> @a.std
=> 4.72228758124704
>> @a.sort
=> [2, 3, 6, 7, 14]
>> @a.rank
=> [2, 1, 3, 4, 5]
>> @a.order  
=> [2, 1, 3, 4, 5]
>> @a.quantile  
=> [2, 3, 6, 7, 14]
>> @a.cum_sum
=> [3.0, 5.0, 11.0, 18.0, 32.0]
>> @a.cum_prod
=> [3.0, 6.0, 36.0, 252.0, 3528.0]
>> @a.cum_max
=> [3, 3, 6, 7, 14]
>> @a.cum_min
=> [3, 2, 2, 2, 2]
>> # And some methods between lists
?> cor(@a,@b)
=> 0.755599551729267
>> # This is the max of each pair (or set)
?> p_max(@a,@b)
=> [4, 6, 6, 7, 19]
>> # These take an arbitrary sized list
?> p_max(@a,@b, [1,2,3,4,5])
=> [4, 6, 6, 7, 19]
>> # And the min
?> p_min(@a,@b)
=> [3, 2, 2, 1, 14]
>> p_min(@a,@b, [1,2,3,4,5])
=> [1, 2, 2, 1, 5]
>> # Finally, some methods out in the wild for general use:
?> max(1,2,3,4,5)
=> 5
>> min(1,2,3,4,5)
=> 1
>> product(1,2,3,4,5)
=> 120.0
>> to_pairs(@a,@b) { |a, b| a * b }
=> [12, 12, 12, 7, 266]
>> sum_pairs(@a,@b) { |a, b| a * b }
=> 309.0

[sirb]$: bin/sirb
Loading sirb (Statistics + Irb: 0.6.3)
>> set :add, lambda {|x,y| x + y}, "This is my general-use adder, that adds a set of numbers"
=> "add added"
>> add 1, 2, 3
=> 6
>> sirb_help :add
* add (takes 2 arguments, returns a single value--reduce function)
  This is my general-use adder, that adds a set of numbers
  Source: |x,y| x + y
=> nil
>> sirb_help

This is Irb, with some extra libraries and commands loaded.
You have loaded the following libraries:
  narray
  rgl
  matrix
  rnum
  set
  statisticus
  rbtree
  rubygems
  command runner
  enumerable statistics
  general statistics
  mathn

You have setup the following commands:

* add (takes 2 arguments, returns a single value--reduce function)
  This is my general-use adder, that adds a set of numbers
  Source: |x,y| x + y

You can store a command like this:
        set :command_name, lambda{|list params| command }, "Optional description"

>> commands
=> [:add]
>> remove_command :add
=> #<Sirb::Runner:0x23df728 @block=proc {|x,y| x + y}, @name="add", @description="This is my general-use adder, that adds a set of numbers">
>> commands
=> []

Sometimes it is useful to have a bit of functional flair when implementing equations. I’ve slurped up a bit of that and included it:

.... Fill these in

I’m generally dissatisfied with quantile, which I think matches the way that R does their quantile. However, R is inconsistent, and the community has inconsistent interpretations of how to handle that, so I just used R as the standard for today.

I only mix this into Array right now. I need to work through the various other classes that I use, and see if this will mix into all of those: Vector, Matrix, Hash, RBTree, etc.

I am a little uncomfortable with my stored procedures library. I want to review it, because it so aggressively walk on Proc. There are other ideas out there that I will want to review when I get the chance.

• All the libraries are optional, but every time you log in, you will get a list of libraries that you may want to install.

• sudo gem install davidrichards-sirb

(The MIT License)

Copyright © 2009 David Richards

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the ‘Software’), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED ‘AS IS’, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.