Even faster CNNs: exploring the new class of winograd algorithms
Convolutional Neural Networks (CNNs) are compute-intensive, with increasingly complex architectures. Learn how the new class of Winograd Algorithms make CNNs faster than before, allowing implementation of workloads like classification and recognition on low-power, Arm-based platforms.
Watch videoCompute Library: optimizing computer vision and machine learning on Arm
Explore industry use cases in which the adoption of optimized low-level primitives for Arm processors enabled improved performance and optimal use of heterogeneous system resources.
Watch videoJump-start machine learning projects with CMSIS-NN on NXP i.MX RT
Learn how to use Arm NN and CMSIS-NN to develop efficient neural network applications for Cortex-M devices. Explore how to use i.MX RT processors with CMSIS-NN to run applications like keyword spotting.
Watch videoAccelerate machine learning using Compute Library and HiKey 960
Learn how to run AI and ML applications on a mobile development platform.
Watch video
Machine learning on deeply embedded and resource-constrained end nodes for the IoT
Learn how hardware technology and software libraries help developers implement tasks like voice recognition without connecting to the cloud.
Watch videoHow to use the MATRIX Creator to unleash the power of voice recognition
Learn how to use the MATRIX Creator to unleash the power of voice recognition. Deploy the Snips voice assistant on the MATRIX Creator using the MATRIX Core programming layer in JavaScript.
Watch videoHow to do edge computing everywhere with a mobile developer workstation
Learn how to do edge computing with a solar-powered mobile developer workstation and extreme edge computing box with 96Boards and miniNodes.
Watch videoEnabling industry 4.0 with NXP's MCU-based machine learning solution for power-conscious end nodes
Learn how to make the most of NXP’s Arm Cortex-M33-based MCU for your power-conscious machine learning application.
Watch videoRunning and profiling Arm NN on the HiKey 960
Learn how to create Linux applications that load TensorFlow trained Neural Network models, run them on Arm Cortex-A CPUs and Mali GPUs, and profile application performance with Arm Streamline.
Watch videoImage recognition on Arm Cortex-M with CMSIS-NN
Learn how to perform real-time image recognition on a low-power Arm Cortex-M7 processor using the Arm CMSIS-NN library.
Watch videoScalable ML acceleration with ONNX Runtime
Technical session delivered by Manash Goswami from Microsoft demonstrates how the ONNX Runtime Execution Providers can run on multiple hardware configurations using a single API.
Watch videoRunning AI and Neural networks on microcontrollers made simple with the STM32Cube.AI
Markus Mayr from STMicroelectronics discusses the STM32Cube.AI toolbox which generates optimized code to run neural networks on the STM32 Arm Cortex-M based microcontrollers
Watch videoMaximize the performance of your ML platform with Arm
Explore how to use Arm’s ML software libraries and tools optimized for Arm endpoint devices.
Watch videoBuilding noise-immune speech interfaces for IoT
Chris Rowen, CEO at BabbleLabs, covers the Clear Command software subsystems that run on Arm Cortex-M processors.
Watch videoExtending machine learning to industrial IoT applications at the edge with AWS
Ian Perez Ponce from AWS discusses real-world use case trends for Industrial IoT (IIoT) applications.
Watch videoSwim builds AI models and predicts in real-time
Dr. Simon Crosby, CTO at Swim.ai, shows how Swim automatically builds, runs and manages scalable, distributed, intelligent dataflow pipelines
Watch videoDesign and optimize computer vision applications on Arm
Dr. Nitin Gupta from Dori AI covers what to consider when trying to annotate, train, optimize, benchmark, deploy, and monitor a computer vision ML application.
Watch videoNext generation machine learning for mobile and embedded platforms
Dr. Rajen Bhatt from Qeexo shows the commercial uses of Qeexo’s engine and suite of tools for optimizing models.
Watch videoAI workflow for large scale deployment of far-edge ML devices
Kabir Manghnani from Shoreline IoT discusses simultaneous training and deployment of ML models trained specifically for the sensors they operate on.
Accelerating ML inference on Raspberry Pi With PyArmNN
Create and optimize on-target run-time performance for advanced ML solutions using the Au-Zone Technologies DeepView ML Toolkit, across a broad spectrum of Arm IP.
Watch videoThe benefits of in-vehicle sensor fusion for autonomous vehicles
Learn how EyerisNet and Arm NN Inference Engine can help you optimize safety and comfort to provide the best in-cabin performance in automotive applications.
Watch videoCommunity Forums
| Answered | How to make Ethos-U NPU work on an ARM Cortex-A + Cortex-M processor? | 0 votes | 4995 views | 22 replies | Latest 2 days ago by alisonw | Answer this |
| Suggested answer | Pytorch framework for Arm NN (CMSIS) | 0 votes | 4314 views | 5 replies | Latest 12 days ago by BCJuan | Answer this |
| Not answered | Top-k scoring across multiple dimensions | 0 votes | 140 views | 0 replies | Started 17 days ago by John Smith | Answer this |
| Suggested answer | Re-build tensorflow lite model in cmsis-nn | 0 votes | 2526 views | 3 replies | Latest 20 days ago by Judd | Answer this |
| Answered | How to make Ethos-U NPU work on an ARM Cortex-A + Cortex-M processor? Latest 2 days ago by alisonw | 22 replies 4995 views |
| Suggested answer | Pytorch framework for Arm NN (CMSIS) Latest 12 days ago by BCJuan | 5 replies 4314 views |
| Not answered | Top-k scoring across multiple dimensions Started 17 days ago by John Smith | 0 replies 140 views |
| Suggested answer | Re-build tensorflow lite model in cmsis-nn Latest 20 days ago by Judd | 3 replies 2526 views |