OpenCog Book / Manual

[nfigl...@gmail.com]

Hi

I feel very attracted to OpenCog, but I find it difficult to follow. The wiki has useful information, but is hard to put hands on work without practical examples of real world scenario usages.
When a book for mortals is going to be published? Are there some tutorials with real world samples?

Thanks.
Nicolas Iglesias

[Cosmo Harrigan]

Hi Nicolas,

An introductory summary of the system is available here:

http://goertzel.org/papers/CogPrime_Overview_Paper.pdf

Further aspects are detailed in the wiki and in the README files in various folders in the code repository.

A very extensive book titled "Engineering General Intelligence" is also forthcoming that will discuss the design in greater detail.

There are also some examples and tutorials available, although more should be implemented in the future. If you can reply with what parts of the system you are most interested in, perhaps someone can point you in the right direction.

Thanks,

Cosmo

--
You received this message because you are subscribed to the Google Groups "opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencog+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
Visit this group at http://groups.google.com/group/opencog.
For more options, visit https://groups.google.com/d/optout.

[nfigl...@gmail.com]

Thanks Cosmo, I already started to read that PDF while ago, very interesting btw.

I am willing to develop some sort of intelligent entity (call it a robot, or intelligent software), by merging OpenCV (Computer Vision) and OpenCog. I don't have any mathematical background, but I am a developer since 10 years ago and Artificial Intelligence is my passion. I've built some Computer Vision software, and right now, i'd like to use OpenCog to apply some reasoning on my projects. I think both libraries could work together very nicely. I don't have a strict plan now, I try to find some free time to work and do research; i'm willing to try Raspberry + OpenCog + OpenCv as the main parts of a personal project.

¿Are there any experience on building OpenCog for ARM processors?

I must admit I'm still confused and do not know the specific functionality for all the OpenCog modules, so, what's the best approach for me right now?

Thanks and regards,
Nicolas

[David Hart]

Hi Nicolas,

DeSTIN is the OpenCog component that interfaces with OpenCV. It has yet to be integrated with the OpenCog CogServer or AtomSpace, so there's plenty of interesting work to be done! https://github.com/opencog/destin http://wiki.opencog.org/w/DeSTIN

No porting efforts have been made to run OpenCog on ARM (armhf or AArch64). In any case, the Raspberry Pi CPU is too weak and memory constrained to run OpenCog usefully. A typical robot vision research effort would involve Wi-Fi streaming OpenCV output to a capable laptop computer (preferably installed with Ubuntu) running OpenCog. The Intel MinnowBoard *might* run OpenCog with acceptable performance for robot vision research.

Others on the list can provide more pointers, depending on where you'd like to focus effort.

-dave

-- 
David Hart
OpenCog Foundation

[Dave Xanatos]

In light of the significant enhancements to the Raspberry Pi in version 3 now, I'd like to re-visit this thread. I am currently running a RPi 3, with Raspbian Jessie, Python 3.4, OpenCV 3.2.0 and support code such as NumPy, SciPy, SKLearn, Swig, etc., doing face detection and recognition, and responding in real-time to visually recognized instances via espeak, and controlling servos via SPI through the GPIOs to control eye movement to maintain "eye contact" with the detected face's eyes. In addition, I am running CMUSphinx/PocketSphinx for speech recognition (still needs significant training). None of this is taxing my Pi significantly. I just discovered OpenCog while searching for an AI framework to pull all my cognitives together and create a "presence", with the goal of creating a reasonably functional conversational interactivity to go along with the CV, hearing and Speech components. Interested in hearing from folks who may have done, or be in process with this. Thanks.

[Nil Geisweiller]

I could build almost all OpenCog components under my RPi3, with the
exception of MOSES cause it took too much RAM, but configuring gcc to
use less RAM should work. I used Ubuntu MATE as distro, I see no reason
it wouldn't work under Raspbian.

Nil

[Dave Xanatos]

Thank you - off to download & install now!  I'll post my results and experience.

Dave

[Dave Xanatos]

One more quick question - is OCTool the proper method for installing on Raspbian (it specifies Ubuntu quite directly)?  

I also found an "installation script" that suggests different dependencies, etc., between Ubuntu and Debian/Raspbian - I have attached that script, although I'm not planning on using it but in reading it and ensuring I am wgetting/cloning/building/making things with the proper files in the proper order.

Thanks again,

Dave

On Wednesday, March 1, 2017 at 3:25:31 AM UTC-5, Nil wrote:

[Linas Vepstas]

Hi Dave

Most of the sensory integration code for OpenCog is in the Hanson Robotics github repos. You seem to already have some ideas about that and it would be interesting to talk about that. I (we?) want to build a fully thinking and sensing robot.  

Re raspberry pi someone else was tellinge they started a port and I guess I have to push them a bit.

Later, Linas

On Wednesday, March 1, 2017, Dave Xanatos <xan...@xanatos.com> wrote:

In light of the significant enhancements to the Raspberry Pi in version 3 now, I'd like to re-visit this thread.  I am currently running a RPi 3, with Raspbian Jessie, Python 3.4, OpenCV 3.2.0 and support code such as NumPy, SciPy, SKLearn, Swig, etc., doing face detection and recognition, and responding in real-time to visually recognized instances via espeak, and controlling servos via SPI through the GPIOs to control eye movement to maintain "eye contact" with the detected face's eyes.  In addition, I am running CMUSphinx/PocketSphinx for speech recognition (still needs significant training).  None of this is taxing my Pi significantly.  I just discovered OpenCog while searching for an AI framework to pull all my cognitives together and create a "presence", with the goal of creating a reasonably functional conversational interactivity to go along with the CV, hearing and Speech components.  Interested in hearing from folks who may have done, or be in process with this.  Thanks.

--
You received this message because you are subscribed to the Google Groups "opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencog+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/41122576-47d7-45b8-b944-55b89e5eb2b7%40googlegroups.com.

[Nil Geisweiller]

On 03/01/2017 08:44 PM, Dave Xanatos wrote:
> One more quick question - is OCTool the proper method for installing on
> Raspbian (it specifies Ubuntu quite directly)?

No idea as I've never used OCTool, I suppose ultimately that would be
the goal.

Nil

> --
> You received this message because you are subscribed to the Google
> Groups "opencog" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to opencog+u...@googlegroups.com

> <mailto:opencog+u...@googlegroups.com>.

> To post to this group, send email to ope...@googlegroups.com

> <mailto:ope...@googlegroups.com>.

> Visit this group at https://groups.google.com/group/opencog.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/opencog/5c8cc559-bd09-4a69-b497-07c0b716a785%40googlegroups.com
> <https://groups.google.com/d/msgid/opencog/5c8cc559-bd09-4a69-b497-07c0b716a785%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Dave Xanatos]

Thank you.  Rather than jump the gun and have to clean up several failed install attempts (not to mention potentially clobbering and breaking my current existing build)...  perhaps I should pose this question to the top level of the group with the proper subject line :)  I am just very anxious to get this running and start trying some ideas I have.

Dave

[supahacka]

"The Raspberry Pi 3 comes nowhere close to the Jetson TX2 in performance."

http://www.computerworld.com/article/3177992/robotics/nvidias-pascal-powered-jetson-tx2-computer-blows-away-raspberry-pi.html

[Ben Goertzel]

The cost of the Jetson makes it unappealing for many envisioned
projects, e.g. the low cost open-source robot kits my colleagues and I
are now selling to African universities (via iCog Makers, a spinoff of
iCog Labs in Addis Ababa)...

But it does look cool ;)

> --
> You received this message because you are subscribed to the Google Groups
> "opencog" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to opencog+u...@googlegroups.com.
> To post to this group, send email to ope...@googlegroups.com.

> Visit this group at https://groups.google.com/group/opencog.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/opencog/96cef80e-ba3e-c03c-2823-c217230048ac%40gmail.com.

>
> For more options, visit https://groups.google.com/d/optout.



--

Ben Goertzel, PhD
http://goertzel.org

“Our first mothers and fathers … were endowed with intelligence; they
saw and instantly they could see far … they succeeded in knowing all
that there is in the world. When they looked, instantly they saw all
around them, and they contemplated in turn the arch of heaven and the
round face of the earth. … Great was their wisdom …. They were able to
know all....

But the Creator and the Maker did not hear this with pleasure. … ‘Are
they not by nature simple creatures of our making? Must they also be
gods? … What if they do not reproduce and multiply?’

Then the Heart of Heaven blew mist into their eyes, which clouded
their sight as when a mirror is breathed upon. Their eyes were covered
and they could see only what was close, only that was clear to them.”

— Popol Vuh (holy book of the ancient Mayas)

[Dave Xanatos]

That does indeed look sweet, but $570 a pop is pretty steep, for now.  But if I can get even a fraction of my planned system running ok in a Pi cluster (proof of concept), I might consider the Jetson TX2 as a far more capable framework.  Thanks for the link.