Shen and a functional ADT with vectors

[Antti Ylikoski]

We have seen in this group a functional ADT done with Shen tuples, and

one implemented with lists.  I made an example of a functional Shen

program with ADTs implemented with Shen vectors.  The demo program is

in the attachment mccpv.shen.  (It is the McCulloch--Pitts neuron

program that I mentioned earlier.)

A sample run is below.

------------------------------------------------------------

antti@antti-HP-630-Notebook-PC:~/ShenADT$ ./Shen

Shen, copyright (C) 2010-2015 Mark Tarver

www.shenlanguage.org, Shen 19.2

running under Common Lisp, implementation: CLisp

port 1.9 ported by Mark Tarver

(0-) (load "mccpv.shen")

type#aneuron

mk-aneuron

pNN

pNDoc

pInp

pWts

pActL

pNOut

sNN

sNDoc

sInp

sWts

sActL

sNOut

type#aneuron-net

<"McCullough-Pitts neuron" "Luger textbook 6th ed. Exercise #1, Ch. 11 neuron" [(@p 0 "neuron-input") (@p 0 "neuron-input") (@p 1 "bias")] [(@p -1 "neuron-input") (@p 1 "neuron-input") (@p 1 "bias")] 0 0>

transfer-function

activation-level-aux

activation-level-aux-h

thres-f

ann-demo

dn

dn

run time: 1.2120000086724758 secs

loaded

(1-) (ann-demo)

Give A: 1

Give B: 1

Neuron name:

McCullough-Pitts neuron

Neuron documentation:

Luger textbook 6th ed. Exercise #1, Ch. 11 neuron

Neuron inputs:

[(@p 1 "user-input") (@p 1 "user-input") (@p 1 "bias")]

Neuron weights:

[(@p -1 "neuron-input") (@p 1 "neuron-input") (@p 1 "bias")]

Neuron activation:

0

Neuron output:

0

Output of neuron: 1

[]

(2-) (ann-demo)

Give A: -1

Give B: -1

Neuron name:

McCullough-Pitts neuron

Neuron documentation:

Luger textbook 6th ed. Exercise #1, Ch. 11 neuron

Neuron inputs:

[(@p -1 "user-input") (@p -1 "user-input") (@p 1 "bias")]

Neuron weights:

[(@p -1 "neuron-input") (@p 1 "neuron-input") (@p 1 "bias")]

Neuron activation:

0

Neuron output:

1

Output of neuron: 1

[]

(3-) (ann-demo)

Give A: 1

Give B: -1

Neuron name:

McCullough-Pitts neuron

Neuron documentation:

Luger textbook 6th ed. Exercise #1, Ch. 11 neuron

Neuron inputs:

[(@p 1 "user-input") (@p -1 "user-input") (@p 1 "bias")]

Neuron weights:

[(@p -1 "neuron-input") (@p 1 "neuron-input") (@p 1 "bias")]

Neuron activation:

0

Neuron output:

1

Output of neuron: -1

[]

(4-) (QUIT)

antti@antti-HP-630-Notebook-PC:~/ShenADT$

------------------------------------------------------------

I had a problem with interfacing the input functions with the Shen

type system: With the OS Shen, the type system generates an error if

one calls (input) or (lineread), and the (read) returns an object of

type unit, which object I cannot assign into a number with the

let ...) form.

========================================

We have here seen that the same abstract datatype, the artificial

neuron, has been implemented:

I) with the Shen tuples; and,

II) with the Shen lists; and,

III) with the Shen vectors; and,

IV) it could be implemented with a dispatching lambda (the author will

leave this one for the reader as an exercise).  Interestingly, calling

a lambda expression and executing a Shen (cases ...) could be cheaper

than eg taking apart a list and replacing one of its elements.

And all this has been done functionally, with the support of the type

system of Shen.

Indeed, the ADTs are a nice thing to have!

yours, A. J. Y.