1f0966c · opencog-agents

Changed files (3) +25 -13

README.md +13 -1

…	…	@@ -3,7 +3,7 @@ Atomese Agents
3	3	Experiments and prototyping for interactive learning agents.
4	4
5	5	### Status
6		Version 0.0.0 -- Extremely early prototyping stage.
	6	Version 0.0.1 -- Extremely early prototyping stage.
7	7
8	8	### Overview
9	9	The [Opencog Learn](https://github.com/opencog/learn) project showed
…	…	@@ -46,3 +46,15 @@ Background reading: the various PDF and LyX files located at:
46	46
47	47	Additional general details and concepts are described in scattershot
48	48	locations, in diary entries, or nowhere at all.
	49
	50	### Action-Perception
	51	The demos here are built on top of the perception-action framework
	52	developed in the [Sensory project](https://github.com/opencog/sensory).
	53	The code here adds some non-trivial processing steps on top of the
	54	basic sensori-motor API provided there. Be sure to go through the
	55	examples there.
	56
	57	For examples in this project, see the [examples](./examples) directory
	58	here.
	59
	60	--------------

examples/README.md +4 -0

…	…	@@ -0,0 +1,4 @@
	1	Demos and Examples
	2	==================
	3
	4	Only one, right now: `pair-count.scm`. Hop to it.

examples/pair-count.scm +8 -12

…	…	@@ -1,20 +1,16 @@
1	1	;
2	2	; pair-count.scm -- Example demo of assembling a word-pair counter.
3	3	;
4		; Prototype hand-built agent capable of counting things observed
5		; in the external environment.
	4	; The generic goal is to be able to observe the external world, and to
	5	; then count things that "happen at the same time". The demo here counts
	6	; word pairs, observed to occur in sentences obtained from a text file.
6	7	;
7		; Demos counting of word-pairs observed by looking at a text file.
8		; The Sensory API allows easy generalization to other text sources,
9		; so working with files is OK for now.
	8	; The demo builds up a processing pipeline, step by step, verifying that
	9	; everythig works at each stage. The result is a carefully-crafted
	10	; counting pipeline. An eventual goal of the sensory project is to
	11	; auto-generate these kinds of pipelines. This demo is meant to
	12	; illustrate a non-trivial pipeline.
10	13	;
11		; The dataflow pipeline is hand-crafted. (The sensory API is supposed
12		; to eventually auto-build these pipelines, but that code is not working
13		; yet.) You'll notice that the pipeline below is ... complicated.
14		;
15		; The "rest of the demo", where we do something with the word pairs,
16		; is in "generate.scm".
17
18	14	(use-modules (opencog) (opencog exec) (opencog persist))
19	15	(use-modules (opencog nlp) (opencog nlp lg-parse))
20	16	(use-modules (opencog sensory))