Akida™ PCIe Board - What have you done & how did you do it?

[JK200SX]

I've started this new thread to collectively gather all the good work that has been done by members in getting the AKIDA PCIE board up and running. We could perhaps use it as a knowledge base to share ideas and forum members can help each other in potential use cases that the may have for the kit.

I've just ordered a kit, through work, as a development project initially to test it out as a potential use case to supersede the existing vision systems we have in place. (I may need peoples help in getting it going )

 

[Fact Finder]

Great idea I have not ordered a board as it would be wasted on me but I have promised one to my son when he gets to the stage of needing one.

In anticipation of someone starting this thread and it becoming a successful group of enthusiasts I have commenced discussions with BRN about supporting it with for example a technical assistance contact point.

So good luck to you and any others who join in.

FF

AKIDA BALLISTA

 

[Equitable]

A board would be well beyond my abilities but I would love to read on here what other more technically-minded people can do with it. Cheers.

 

[JK200SX]

The application I want to try the AKIDA PCIE board on is a high speed container manufacturing line. The containers are injection moulded and have in mould labeling that shows what the product is. The aim is to ensure that there is no mixed up labels in the batch production run and because the containers can come of the line in a different rotational direction the label wont necessarily be presented to the camera in the exact same orientation - and this is where conventional/current vision systems fail. Therefore I'm keen to see if I can get AKIDA to accurately do the detection.

Although I'm an engineer with a materials/ chemical/ manufacturing forte, and a fairly good grasp of computers, both hardware and software, I have no skills in python programming and this is where I may need to call on help from the brainstrust here.

 

[Techinvestor17]

Mines hanging on my wall in a frame, no idea how to get it going but would be interested in trying to get the dev software to work. Can you run it on Mac or only windows?

 

[GrandRhino]

Mines hanging on my wall in a frame, no idea how to get it going but would be interested in trying to get the dev software to work. Can you run it on Mac or only windows?

Would love to see a photo!

 

[Techinvestor17]

Would love to see a photo!

Was going to put I’m not sure how to upload photos anymore

 

[JDelekto]

After receiving my board, I purchased a modest, refurbished PC for about $580 (USD) to run Ubuntu 20.04 as I typically work on Windows systems. I was able to get a customized Dell OptiPlex 7050 SFF (Small Form Factor) Desktop, which looks similar to BrainChip's shuttle PC.

The purchase provided the following specs:

• Intel Core i7-7700K, 4-Core, 4.2GHz
• 16GB (2x8) DDR4 RAM
• 256GB SATA SSD
• Nvidia Quadro K600 Low Profile, 1GB DDR3 VRAM
• Slim DVD+RW Drive

So for the PC and the Akida PCIe board, I spent roughly ~1000, which is about 1/5th of the cost of the shuttle PC. Yes, unfortunately, I contributed to their lower revenues this past quarter.

I spent a weekend installing/configuring Linux, installing development tools such as Python 3 (the version required for MetaTF), Visual Studio Code with associated add-ons, and several other Open Source apps I use under Windows.

I installed the BrainChip PCIe card, and it is useful to note that the mounting bracket and screw were useless to me, as the bracket was too tall for the small form-factor case. Instead, I just seated the card firmly. The card is low-profile enough that it is firmly seated without any play.

I provided my serial number on BrainChip's site to get access to the PDF with the driver installation instructions. I was able to install the drivers as documented and run the tests to ensure the card was working. To my surprise, the card has a tiny bright-blue LED light when powered on.

What I have yet to do before I consider the system complete:

• Acquire a Web and set up a Web camera.
• Finishing a course or two for TensorFlow on Udemy.
• Play around with the existing MetaTF examples.

The first project I have planned is a hobby project incorporating the use of the Emotiv Insight Brain-Computer Interface (BCI) which uses electroencephalography (EEG) to train and recognize commands that can be personalized for the individual using it. Several years ago, a Community SDK was provided (I will need to sift through all the forked branches) for developers to write custom interfaces to the device.

The initial goal is to figure out how to take that raw EEG data and determine how to provide that information to Akida's neural fabric to make use of its one-shot (or multi-shot) learning. The data collection application would consist of showing the user training the chip to read a command (which would be provided by a visual representation) while they concentrate on how they would perform it. Simple primitives like Stop, Go, Right, and Left would be sampled, then the user could "think" about these commands and the detected results from the training data would be displayed.

I should note that the Emotiv software itself provides a means of gathering EEG information and can play it back. Think of it like "Dragon Naturally Speaking", but instead of speech recognition, it would be recognizing brain activity. What I hope to accomplish is a proof of concept for Smart Edge devices that can be individually trained and controlled by a user on-demand.

A secondary, but future project would be to analyze the PCIe driver for which BrainChip provides the Linux source code and create a Windows PCIe driver. While it is fun to play with Linux once in a while, I'm mainly a Windows developer. It's my comfort zone for being more productive.

 

[JDelekto]

The application I want to try the AKIDA PCIE board on is a high speed container manufacturing line. The containers are injection moulded and have in mould labeling that shows what the product is. The aim is to ensure that there is no mixed up labels in the batch production run and because the containers can come of the line in a different rotational direction the label wont necessarily be presented to the camera in the exact same orientation - and this is where conventional/current vision systems fail. Therefore I'm keen to see if I can get AKIDA to accurately do the detection.

Although I'm an engineer with a materials/ chemical/ manufacturing forte, and a fairly good grasp of computers, both hardware and software, I have no skills in python programming and this is where I may need to call on help from the brainstrust here.

While I have used very little Python in the past hacking together a few Blender scripts, I too am on a journey to learn it. I work with several different languages, so fortunately I can pick up the basics quickly, however, to be proficient at them and use best practices does take a bit more time.

I am several days into a Udemy course, 100 Days of Code: The Complete Python Pro Bootcamp for 2022, which I can recommend, but with a condition. I like this course because they allow you to sign up on a site (actually two, but you don't necessarily realize that until you get into a later lesson) that allows you to code the exercises live, where they are compiled and the results are validated.

What I like about the "hands-on" approach is that I find it a better way to learn a new language Caveat Emptor: the exercises are contrived examples and you will not learn how to write "professional" quality Python code. To be fair, I haven't reached the end of the course so I cannot speak to the quality of later training. However, at this point, I can confidently say that if I wrote code using what I learned so far, the experts would know that I am a novice Python developer.

For hacking on fun projects though, it should be more than sufficient.

 

[hotty4040]

Would love to see a photo!

Me to GR, I would frame the photo and pretend I'd bought one of the boards, because at this stage of my existence I wouldn't have a clue as to how, why or when I could utilize it.

Besides, I haven't seen my soldering iron for years, or the sucker, that was used extensively, they are still around though somewhere. there probably well rusted out by now. I still do have some other equipment i.e. multimeter, reels of solder, and a container full of all sorts of components, i.e. transistors, resisters, even a variac that's never been used by me, loads of circuit cards etc, but alas, not required now, unfortunately.

Ah, evidence of a former interest and occupation, which I just loved and made a living from. Those days have been numbered now for a while, but do very much miss that challenge and travel opportunities it enabled.

Just content now to contest the share market opportunities that I found a few years ago. Just wish I'd found these on-line share brokers much earlier in my life, because I do still like a good gamble from time to time, and this is a most compelling gamble. I find the stock market and all it's shenanigans so compelling and interesting, especially at this time, Brainchip and this forum, which I'm sort of pretty well addicted to, with a number of stocks I'm with and following. Still on hot crapper, but it's slowly loosing it's interest. There are Gambits a plenty on offer I've found.

Anyway, I await the posting of the Akida PCIE card photo, can't wait to frame that for posterity purposes.

Akida Ballista comrades, GLTAH.

hotty...

 

[JK200SX]

After receiving my board, I purchased a modest, refurbished PC for about $580 (USD) to run Ubuntu 20.04 as I typically work on Windows systems. I was able to get a customized Dell OptiPlex 7050 SFF (Small Form Factor) Desktop, which looks similar to BrainChip's shuttle PC.

The purchase provided the following specs:

• Intel Core i7-7700K, 4-Core, 4.2GHz
• 16GB (2x8) DDR4 RAM
• 256GB SATA SSD
• Nvidia Quadro K600 Low Profile, 1GB DDR3 VRAM
• Slim DVD+RW Drive

So for the PC and the Akida PCIe board, I spent roughly ~1000, which is about 1/5th of the cost of the shuttle PC. Yes, unfortunately, I contributed to their lower revenues this past quarter.

I spent a weekend installing/configuring Linux, installing development tools such as Python 3 (the version required for MetaTF), Visual Studio Code with associated add-ons, and several other Open Source apps I use under Windows.

I installed the BrainChip PCIe card, and it is useful to note that the mounting bracket and screw were useless to me, as the bracket was too tall for the small form-factor case. Instead, I just seated the card firmly. The card is low-profile enough that it is firmly seated without any play.

I provided my serial number on BrainChip's site to get access to the PDF with the driver installation instructions. I was able to install the drivers as documented and run the tests to ensure the card was working. To my surprise, the card has a tiny bright-blue LED light when powered on.

What I have yet to do before I consider the system complete:

• Acquire a Web and set up a Web camera.
• Finishing a course or two for TensorFlow on Udemy.
• Play around with the existing MetaTF examples.

The first project I have planned is a hobby project incorporating the use of the Emotiv Insight Brain-Computer Interface (BCI) which uses electroencephalography (EEG) to train and recognize commands that can be personalized for the individual using it. Several years ago, a Community SDK was provided (I will need to sift through all the forked branches) for developers to write custom interfaces to the device.

The initial goal is to figure out how to take that raw EEG data and determine how to provide that information to Akida's neural fabric to make use of its one-shot (or multi-shot) learning. The data collection application would consist of showing the user training the chip to read a command (which would be provided by a visual representation) while they concentrate on how they would perform it. Simple primitives like Stop, Go, Right, and Left would be sampled, then the user could "think" about these commands and the detected results from the training data would be displayed.

I should note that the Emotiv software itself provides a means of gathering EEG information and can play it back. Think of it like "Dragon Naturally Speaking", but instead of speech recognition, it would be recognizing brain activity. What I hope to accomplish is a proof of concept for Smart Edge devices that can be individually trained and controlled by a user on-demand.

A secondary, but future project would be to analyze the PCIe driver for which BrainChip provides the Linux source code and create a Windows PCIe driver. While it is fun to play with Linux once in a while, I'm mainly a Windows developer. It's my comfort zone for being more productive.

Instead of a linux based system, can it run on Windows?

 

[GrandRhino]

Me to GR, I would frame the photo and pretend I'd bought one of the boards, because at this stage of my existence I wouldn't have a clue as to how, why or when I could utilize it.

Besides, I haven't seen my soldering iron for years, or the sucker, that was used extensively, they are still around though somewhere. there probably well rusted out by now. I still do have some other equipment i.e. multimeter, reels of solder, and a container full of all sorts of components, i.e. transistors, resisters, even a variac that's never been used by me, loads of circuit cards etc, but alas, not required now, unfortunately.

Ah, evidence of a former interest and occupation, which I just loved and made a living from. Those days have been numbered now for a while, but do very much miss that challenge and travel opportunities it enabled.

Just content now to contest the share market opportunities that I found a few years ago. Just wish I'd found these on-line share brokers much earlier in my life, because I do still like a good gamble from time to time, and this is a most compelling gamble. I find the stock market and all it's shenanigans so compelling and interesting, especially at this time, Brainchip and this forum, which I'm sort of pretty well addicted to, with a number of stocks I'm with and following. Still on hot crapper, but it's slowly loosing it's interest. There are Gambits a plenty on offer I've found.

Anyway, I await the posting of the Akida PCIE card photo, can't wait to frame that for posterity purposes.

Akida Ballista comrades, GLTAH.

hotty...

Sounds a bit like me too at the moment, I used to be a lot more into electronics and tinkering with things but slowly shifting my focus to the big scary world of the share market.

I wonder if BRN decided to sell a framed photo of the chip whether they would sell well, I probably would buy one!

 

[JDelekto]

Instead of a linux based system, can it run on Windows?

I sent an e-mail to support and inquired if they had a driver for Windows, but unfortunately, they only provide PCIe drivers for Ubuntu 18.04 and 20.06.

That's why one of my future goals is to see if I could analyze the source for the Linux PCIe driver (which BrainChip does make available on GitHub) and see if I can create a driver for Windows as well.

I'm not sure if there is some type of "shim" that needs to be written to use this with Python, but that may have to be done as well. However, with a driver, I see no reason why the Akida board couldn't be used with Windows as well.

 

[Fact Finder]

Great idea I have not ordered a board as it would be wasted on me but I have promised one to my son when he gets to the stage of needing one.

In anticipation of someone starting this thread and it becoming a successful group of enthusiasts I have commenced discussions with BRN about supporting it with for example a technical assistance contact point.

So good luck to you and any others who join in.

FF

AKIDA BALLISTA

Just an update. I received a reply and the idea of a Supporters Club has some appeal and is to be discussed so watch this space.
FF

 

[AusEire]

Just an update. I received a reply and the idea of a Supporters Club has some appeal and is to be discussed so watch this space.
FF

Isn't that what we have here? Lol

 

[JK200SX]

Not long before I get to hold some silicon in my hands

 

[JK200SX]

Silicon in hand .....

 

[JDelekto]

Silicon in hand .....

View attachment 6071

I was surprised by how small the board was. I expected it to be the size of a floppy controller from back in the early 1990s, but it turned out to be a munchkin of a card. I guess it just looked bigger in the marketing photos. Then again, someone showing the processor itself compared to their hand was about the size of a thumbnail. Akida's probably better at extrapolating relative sizes than I am.

Now you can see why I wasn't worried about not being able to fit the mounting bracket in my small form-factor PC case as the thing can pretty much hold its own with the PCIe connector and the small cutout just above the ADK1000.

The one thing that I did find strange was that copper-colored wire jumper on the board just below the connectors as you have pictured. It almost looks like it was a modification done after the initial design. Maybe they needed to "patch" something.

Have fun!

 

[JK200SX]

I was surprised by how small the board was. I expected it to be the size of a floppy controller from back in the early 1990s, but it turned out to be a munchkin of a card. I guess it just looked bigger in the marketing photos. Then again, someone showing the processor itself compared to their hand was about the size of a thumbnail. Akida's probably better at extrapolating relative sizes than I am.

Now you can see why I wasn't worried about not being able to fit the mounting bracket in my small form-factor PC case as the thing can pretty much hold its own with the PCIe connector and the small cutout just above the ADK1000.

The one thing that I did find strange was that copper-colored wire jumper on the board just below the connectors as you have pictured. It almost looks like it was a modification done after the initial design. Maybe they needed to "patch" something.

Have fun!

Nah, I’ve just got giant hands

I’ll have a go at installing Ubuntu later today. I have a spare i3-6100 HP Prodesk lying around that I’ll use with the OS on an SSD.

 

[Deleted member 118]

Products - AI Enablement

Find out what Brainchip AI can do for you!

brainchip.com

AI Enablement​

Sensors make digital interaction more straightforward and more meaningful, and smart sensors make edge AI possible.
BrainChip’s AI Enablement Program makes entry to edge AI simple and real. Each of our tiered programs brings you from concept to working prototype with varying model complexity and sensor integration levels.
In addition, our AI experts provide training and support to make the process efficient and smart.

Contact Sales

Basic Prototype
Package​

BrainChip’s Basic Prototype Package includes hardware, a machine learning model from our expansive model zoo, training, and engineering and support from a dedicated BrainChip team. Clients will achieve a basic prototype to propel their enterprise AI journey.

Advanced Prototype
Package​

BrainChip’s Advanced Prototype Package includes hardware, a custom machine learning network, training, and engineering and support from a dedicated BrainChip team. Clients will achieve a custom network trained on their specific dataset and tailored to their target use case.

Functioning Prototype
Package​

BrainChip’s Functioning Prototype Package includes hardware, 2 custom machine learning networks, detailed training, and expansive engineering and support from a dedicated BrainChip team, including a technical project manager.