Learning
Learning to the Top 🕵♂️
|
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
MaxwellAnne
Regular
Brilliant @Bravo !!
Stable Genius
Regular
Another like from Anil. I’m hopeful this works with Brainchip as further in the description it talks about accelerating neural networks.
Not sure how this fits into the equation regarding synopsys ?
Not sure how this fits into the equation regarding synopsys ?
buena suerte :-)
BOB Bank of Brainchip
Great to see your post this morning BL ... welcome back home!!
and can we have our "Iceberg back'
Last edited:
Stable Genius
Regular
alwaysgreen
Top 20
Finally some bloody green!
ISO 9001 - or Total Quality Management - sort of a company wide way of doing things from
This is our mission
This is how we show we are achieving that by constantly looking for possible causes of mis-steps.
uiux
Regular
The MLSoC provides:
• Best-in-class DNN inference efficiency–
500 FPS/W ResNet-50 v1, batch size 1.
• Up to 50 TOPS for neural network
computation.
The MLSoC contains the following major functions:
• Machine learning accelerator (MLA) – providing 50 teraoperations per second (50 TOPS) for neural network computation at 10 TOPS/W.
Computer vision unit (CVU) – consists of a
four-core Synopsys ARC EV74 video processor supporting up to 600 16-bit GOPS.
MLSoC offers heterogeneous cores for processing any ML workload. Quad Arm A65 cores, a Machine Learning Accelerator (MLA) block that provides up to 50 TOPS for ML acceleration along with a Computer Vision Processor (CVP) to address any ML computational needs.
---
It looks like SIMA.ai could be potential licensees
Nothing in current hardware that screams neuromorphic
AC/DC
Member
The last 2 paragraphs are exquisite...and required reading!A lot to consider in this following linked article. If generally accurate the following seems possible:
1. The USA will not become an inflationary mess and once again will become the dominant World economy;
2. Europe will become US focused;
3. China & Russian economies will suffer and cause internal instability and it is likely political change will overtake the present leaderships as it has historically in both countries;
4. Australia will keep shooting itself in the foot economically and politically.
Cheap US Energy to Fuel Renewables Transition - ShareCafe
In an undeniable paradox, the cheapness of its carbon-based energy is making America, the world's most profligate carbon user, an increasingly attractive place to develop renewable energy sources.www.sharecafe.com.au
My opinion only DYOR
FF
AKIDA BALLISTA
"Finally, there is one huge irony here. The cheapness of oil and gas / carbon-based energies is making America – the most profligate carbon-based economy on this earth – an increasingly attractive place to build and expand the renewable energy sources that will, hopefully, make the planet a cooler place in decades to come.
That message is not well understood in Australia."
Bravo
If ARM was an arm, BRN would be its biceps💪!
Check this out.
Synopsys NPX6 NPU IP:
June 13, 2022
Synopsys has released a Neural Processing Unit (NPU), Intellectual Property (IP) core, and toolchain to address the performance needs of the increasingly complicated neural network models in AI Systems on Chip (SoCs). Its new DesignWare ARC NPX6 and NPX6FS NPU IP handle the demands of real-time computing while consuming ultra-low power for AI applications. Additionally, the company's new MetaWare MX development tools provide a full compilation environment with automated neural network algorithm partitioning to maximize resource efficiency for application software development on the newest NPU.
Using the new DesignWare ARC NPX6 and NPX6FS NPU IP, as well as MetaWare MX development toolkits, designers can take advantage of the newest neural network models, satisfying the escalating performance expectations, and faster time-to-market for their next intelligent SoCs. The ARC NPX6 NPU IP family includes numerous products that handle deep learning algorithm coverage, including computer vision tasks such as object recognition, picture quality enhancement, and scene segmentation, as well as larger AI applications such as audio and natural language processing. Individual cores in the design may scale from 4K MACs to 96K MACs for a single AI engine performance of over 250 TOPS and over 440 TOPS with sparsity.
The NPX6 NPU IP contains hardware and software support for multi-NPU clusters of up to eight NPUs with a sparsity of 3500 TOPS. Scaling to a large MAC count is achievable, thanks to advanced bandwidth features in hardware and software, as well as a memory hierarchy (containing L1 memory in each core and a high-performance, low-latency connection to access common L2 memory). For applications that benefit from BF16 or FP16 inside the neural network, an optional tensor floating-point unit is offered.
Image Credits: Synopsys
The MetaWare MX development toolkit provides a software programming environment for application software development, including a neural network software development kit (NN SDK) and virtual model support. The NN SDK automatically turns neural networks trained using popular frameworks such as Pytorch, Tensorflow, or ONNX into NPX-optimized executable code.
The concept is that the NPX6 NPU processor IP may subsequently be used to make a variety of products, ranging from a few TOPS to thousands of TOPS, all of which can be written using a single toolchain.
Moreover, the ARC NPX6FS NPU IP complies with ISO 26262 ASIL D standards for random hardware failure detection and systematic functional safety development flow. The processors have specific safety mechanisms for ISO 26262 compliance and handle the mixed-criticality and virtualization needs of next-generation zonal designs, as well as thorough safety documentation.
The ARC MetaWare MX development toolkit includes a neural network software development kit (SDK), compilers and debugger, virtual platforms SDK, runtimes and libraries, and advanced simulation models. It offers a unified toolchain environment to speed up application development and intelligently divides algorithms among MAC resources for optimal processing. The MetaWare MX development toolkit for safety contains a safety handbook and a safety guide to assist developers to satisfy ISO 26262 criteria and preparing for ISO 26262 compliance testing for safety-critical automotive applications.
To address the increasing performance and complex needs of the AI applications, the NXP NPU IP core offers high performance, scalable real-time AI and neural processing IP with up to 3500 TOPS supporting various neural networks like CNNs, and RNNs/LSTMs, transformers, and recommenders networks.
Further, it reduces the latency through parallel processing of individual layers. Moreover, the high productivity MetaWare MX Development Toolkit supports Tensorflow and Pytorch frameworks and the ONNX exchange format.
For more information related to Synopsys’s NPU IP, register for the session at the Embedded Vision Summit on May 19 where Synopsys will present the concepts regarding the optimization of AI performance and power for the applications of deep learning neural networks.
embeddedcomputing.com
Synopsys NPX6 NPU IP:
- Scalable real-time AI / neural processor IP with the performance of up to 3,500 TOPS supporting CNNs, RNNs/LSTMs, transformers, recommender networks, and other neural networks.
- Power efficiency (up to 30 TOPS/W), is unrivaled in the industry.
- The ARC NPX6 NPU IP family includes numerous products that handle deep learning algorithm coverage, including computer vision tasks such as object recognition, picture quality enhancement, and scene segmentation, as well as larger AI applications such as audio and natural language processing.
Transforming Edge AI with Clusters of Neural Processing Units
By Saumitra Jagdale
Freelance Technology WriterJune 13, 2022
As the field of Artificial Intelligence gains traction, the devices are becoming more computational and power-consuming. Subsequently, the processing load on the edge devices significantly grows with the performance and the complexity of the system architecture. Hence, higher resolution images and more sophisticated algorithms are inculcated in the system which further needs to be optimized with the increasing demand for AI processing to achieve high TOPS performance.
Synopsys has released a Neural Processing Unit (NPU), Intellectual Property (IP) core, and toolchain to address the performance needs of the increasingly complicated neural network models in AI Systems on Chip (SoCs). Its new DesignWare ARC NPX6 and NPX6FS NPU IP handle the demands of real-time computing while consuming ultra-low power for AI applications. Additionally, the company's new MetaWare MX development tools provide a full compilation environment with automated neural network algorithm partitioning to maximize resource efficiency for application software development on the newest NPU.
Using the new DesignWare ARC NPX6 and NPX6FS NPU IP, as well as MetaWare MX development toolkits, designers can take advantage of the newest neural network models, satisfying the escalating performance expectations, and faster time-to-market for their next intelligent SoCs. The ARC NPX6 NPU IP family includes numerous products that handle deep learning algorithm coverage, including computer vision tasks such as object recognition, picture quality enhancement, and scene segmentation, as well as larger AI applications such as audio and natural language processing. Individual cores in the design may scale from 4K MACs to 96K MACs for a single AI engine performance of over 250 TOPS and over 440 TOPS with sparsity.
The NPX6 NPU IP contains hardware and software support for multi-NPU clusters of up to eight NPUs with a sparsity of 3500 TOPS. Scaling to a large MAC count is achievable, thanks to advanced bandwidth features in hardware and software, as well as a memory hierarchy (containing L1 memory in each core and a high-performance, low-latency connection to access common L2 memory). For applications that benefit from BF16 or FP16 inside the neural network, an optional tensor floating-point unit is offered.
Image Credits: Synopsys
The MetaWare MX development toolkit provides a software programming environment for application software development, including a neural network software development kit (NN SDK) and virtual model support. The NN SDK automatically turns neural networks trained using popular frameworks such as Pytorch, Tensorflow, or ONNX into NPX-optimized executable code.
The concept is that the NPX6 NPU processor IP may subsequently be used to make a variety of products, ranging from a few TOPS to thousands of TOPS, all of which can be written using a single toolchain.
Key Features of the NPX6 NPU IP:
- Scalable real-time AI / neural processor IP with the performance of up to 3,500 TOPS supporting CNNs, RNNs/LSTMs, transformers, recommender networks, and other neural networks.
- Power efficiency (up to 30 TOPS/W), is unrivaled in the industry.
- 1-24 cores of a convolution accelerator with an increased 4K MAC/core
- Tensor accelerator that supports the Tensor Operator Set Architecture and allows for variable activation (TOSA)
- Kit for Software Development
- Tools for automatically mixed-mode quantization
- Architecture and software tool characteristics that reduce bandwidth
- Latency is reduced by processing individual layers in parallel.
- DesignWare ARC VPX vector DSPs are seamlessly integrated.
- Productivity is high. Tensorflow and Pytorch frameworks, as well as the ONNX interchange standard, are supported by MetaWare MX Development Toolkit.
Moreover, the ARC NPX6FS NPU IP complies with ISO 26262 ASIL D standards for random hardware failure detection and systematic functional safety development flow. The processors have specific safety mechanisms for ISO 26262 compliance and handle the mixed-criticality and virtualization needs of next-generation zonal designs, as well as thorough safety documentation.
The ARC MetaWare MX development toolkit includes a neural network software development kit (SDK), compilers and debugger, virtual platforms SDK, runtimes and libraries, and advanced simulation models. It offers a unified toolchain environment to speed up application development and intelligently divides algorithms among MAC resources for optimal processing. The MetaWare MX development toolkit for safety contains a safety handbook and a safety guide to assist developers to satisfy ISO 26262 criteria and preparing for ISO 26262 compliance testing for safety-critical automotive applications.
Accelerating Edge AI Applications with Clusters of NPUs
To address the increasing performance and complex needs of the AI applications, the NXP NPU IP core offers high performance, scalable real-time AI and neural processing IP with up to 3500 TOPS supporting various neural networks like CNNs, and RNNs/LSTMs, transformers, and recommenders networks.
Further, it reduces the latency through parallel processing of individual layers. Moreover, the high productivity MetaWare MX Development Toolkit supports Tensorflow and Pytorch frameworks and the ONNX exchange format.
For more information related to Synopsys’s NPU IP, register for the session at the Embedded Vision Summit on May 19 where Synopsys will present the concepts regarding the optimization of AI performance and power for the applications of deep learning neural networks.
Transforming Edge AI with Clusters of Neural Processing Units - Embedded Computing Design
As the field of Artificial Intelligence gains traction, the devices are becoming more computational and power-consuming. Subsequently, the processing load on the edge devices significantly grows with the performance and the complexity of the system architecture. Hence, higher resolution images...
Stable Genius
Regular
SiMa AI.............This could well be another NDA that's popped up out of nowhere.
Anyone else have that feeling.... apart from Lou, for example??
Tech....
I unfortunately don’t know enough about SiMa AI. Hopeful with AM liking it there is a link but there’s no guarantees.
They are relatively new company, founded Nov 2018 so they’ve done well if they’ve got what they’re advertising!
I couldn’t see they had a NN onboard and their website focuses on software. It does include being able to run any network INCLUDING NEURAL NETWORKS therefore I’m thinking it’s a competitor to edge impulse more-so than BRN as it will complement us and help us get to market.
Maybe we can be part of their solution or offering.
Colorado23
Regular
I listened to the afr's how I made this podcast with Aaron Hornlimann. https://www.podbean.com/ea/dir-3pr6t-14fd1ea1
Thought it might be a good opportunity to make Aaron aware of Brainchip if he isn't already.
www.elenium.com
Thought it might be a good opportunity to make Aaron aware of Brainchip if he isn't already.
Elenium - Leading automated self-service solutions company
Elenium is a leading digital identification company. We provide automated self-service solutions to facilitate frictionless movement across a variety of industries with our vision to improve tech effectiveness.
www.elenium.com
Attachments
Fact Finder
Top 20
Social media ‘likes’ do they in and of themselves tell us anything worthy of being used to make an investment decision. The answer is a raging NO.
The use of the ‘like’ button is completely meaningless or so full of subtext that it is still meaningless unless you canvass the person who hit ‘like’ as to why they did so.
I am sure at least once a day just as I do you will read a post here with multiple ideas expressed therein and one of the many ideas will appeal for some reason but the others not so and you give a ‘like’ because either you can’t be bothered posting or are pressed for time or you just don’t post.
Anyway the following linked article gives a little insight into the meaningless world of likes and their subtext. One of the subtext meanings I find interesting is where a ‘like’ is intended to convey good luck with that plan loser.
Have you ever asked yourself could Rob Telson have tried to sell AKIDA to one of these companies and was knocked back so his ‘like’ is intended to convey good luck with your outdated technology you loser:
www.mic.com
So in the absence of something substantive my view is ‘likes’ are meaningless no matter how many are given unless there is something which clearly explains the reason for giving the ‘like’.
My opinion only DYOR
FF
AKIDA BALLISTA
The use of the ‘like’ button is completely meaningless or so full of subtext that it is still meaningless unless you canvass the person who hit ‘like’ as to why they did so.
I am sure at least once a day just as I do you will read a post here with multiple ideas expressed therein and one of the many ideas will appeal for some reason but the others not so and you give a ‘like’ because either you can’t be bothered posting or are pressed for time or you just don’t post.
Anyway the following linked article gives a little insight into the meaningless world of likes and their subtext. One of the subtext meanings I find interesting is where a ‘like’ is intended to convey good luck with that plan loser.
Have you ever asked yourself could Rob Telson have tried to sell AKIDA to one of these companies and was knocked back so his ‘like’ is intended to convey good luck with your outdated technology you loser:
Here's What a "Like" Really Means on Every Social Media Platform
If the rise of social media has taught us anything, it's that humans are lazy and would rather click a button than put a sentiment into words. The ultimate example of this phenomenon? The like. While the deceptively simple purpose of the "like"…
So in the absence of something substantive my view is ‘likes’ are meaningless no matter how many are given unless there is something which clearly explains the reason for giving the ‘like’.
My opinion only DYOR
FF
AKIDA BALLISTA
Similar threads
