Edge-to-Cloud Compute with MxNet

Similar documents
MIOVISION DEEP LEARNING TRAFFIC ANALYTICS SYSTEM FOR REAL-WORLD DEPLOYMENT. Kurtis McBride CEO, Miovision

Machine Learning. Bridging the OT IT Gap for Machine Learning with Ignition and AWS Greengrass

Realtime Object Detection and Segmentation for HD Mapping

Embarquez votre Intelligence Artificielle (IA) sur CPU, GPU et FPGA

Transforming Transport Infrastructure with GPU- Accelerated Machine Learning Yang Lu and Shaun Howell

Deep Learning for Developers

Deep Learning: Transforming Engineering and Science The MathWorks, Inc.

NVIDIA PLATFORM FOR AI

Accelerating your Embedded Vision / Machine Learning design with the revision Stack. Giles Peckham, Xilinx

A NEW COMPUTING ERA. Shanker Trivedi Senior Vice President Enterprise Business at NVIDIA

Deploying Deep Learning Networks to Embedded GPUs and CPUs

EMBEDDED MACHINE VISION

Edge of Tomorrow Deploying Collaborative Machine Intelligence to the Edge

Nvidia Jetson TX2 and its Software Toolset. João Fernandes 2017/2018

S INSIDE NVIDIA GPU CLOUD DEEP LEARNING FRAMEWORK CONTAINERS

Deep Learning on AWS with TensorFlow and Apache MXNet

A NEW COMPUTING ERA JENSEN HUANG, FOUNDER & CEO GTC CHINA 2017

Advancing State-of-the-Art of Autonomous Vehicles and Robotics Research using AWS GPU Instances

Installation guide. Step1. Install the OS for Raspberry Pi 3. Go to to download Raspbian to the PC.

NVIDIA DATA LOADING LIBRARY (DALI)

An introduction to Machine Learning silicon

Serverless Predictions at Scale

Voice, Image, Video : AI in action with AWS. 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.

What s inside: What is deep learning Why is deep learning taking off now? Multiple applications How to implement a system.

Deep learning in MATLAB From Concept to CUDA Code

Inference Optimization Using TensorRT with Use Cases. Jack Han / 한재근 Solutions Architect NVIDIA

Fast Hardware For AI

Introduction to Deep Learning in Signal Processing & Communications with MATLAB

Intelligent Video Analytics for Urban Management

Embedded GPGPU and Deep Learning for Industrial Market

An Introduction to Developing for Cisco Kinetic

S7348: Deep Learning in Ford's Autonomous Vehicles. Bryan Goodman Argo AI 9 May 2017

NVIDIA GPU CLOUD DEEP LEARNING FRAMEWORKS

Fog Computing Design Principles for Industrial IoT Applications. Sudhanshu Gaur Director, Industrial IoT R&D Hitachi America, Ltd.

Small is the New Big: Data Analytics on the Edge

Managing Deep Learning Workflows

AUTONOMOUS DRONE NAVIGATION WITH DEEP LEARNING

Demystifying Deep Learning

Hmax Documentation Documentation

GPU Coder: Automatic CUDA and TensorRT code generation from MATLAB

Implementing Deep Learning for Video Analytics on Tegra X1.

End to End Optimization Stack for Deep Learning

Presenter: Felix Goldberg, Ph.D. Chief AI Scientist Artificial Intelligence Cart AIC

Cluster of Web-Servers with AWS

Intelligent IoT Edge Computing Platform. AWS Greengrass QIoT QuAI

Raspberry Pi Using Open CV which Has The Installing,making Programs And Performance

Lecture 12: Model Serving. CSE599W: Spring 2018

Data Analytics at Logitech Snowflake + Tableau = #Winning

Accelerating intelligence at the edge for embedded and IoT applications

Introduction to Internet of Things Prof. Sudip Misra Department of Computer Science & Engineering Indian Institute of Technology, Kharagpur

CUDNN. DU _v07 December Installation Guide

Raspberry PI 'How-To' Series

Opportunities for ML Analytics at the Sensor Endpoint

AutoTVM & Device Fleet

Easier than Ever: An Introduction for Building End to End IoT Solutions

Bring Intelligence to the Edge with Intel Movidius Neural Compute Stick

Embedded Linux Conference San Diego 2016

Characterization and Benchmarking of Deep Learning. Natalia Vassilieva, PhD Sr. Research Manager

HOW TO BUILD A MODERN AI

Lessons Learned: Building Scalable & Elastic Akka Clusters on Google Managed Kubernetes. - Timo Mechler & Charles Adetiloye

Intelligent Edge Computing and ML-based Traffic Classifier. Kwihoon Kim, Minsuk Kim (ETRI) April 25.

THE LEADER IN VISUAL COMPUTING

Demystifying Deep Learning

In partnership with. VelocityAI REFERENCE ARCHITECTURE WHITE PAPER

BUILDING A GPU-FOCUSED CI SOLUTION

Oracle Machine Learning Notebook

Artec Leo. A smart professional 3D scanner for a next-generation user experience

Is your IT Infrastructure Ready for Machine Learning & Artificial Intelligence?

Artec Leo. A smart professional 3D scanner for a next-generation user experience

Accelerate AI with Cisco Computing Solutions

Artec Leo. A smart professional 3D scanner for a next-generation user experience

DIY Security Camera using. Intel Movidius Neural Compute Stick

CNN optimization. Rassadin A

USING NGC WITH GOOGLE CLOUD PLATFORM

AWS DeepLens Workshop: Building a Computer Vision App

Face Recognition Documentation

P I X E V I A : A I B A S E D, R E A L - T I M E C O M P U T E R V I S I O N S Y S T E M F O R D R O N E S

BAYESIAN GLOBAL OPTIMIZATION

Building Applications with IOx

DGX UPDATE. Customer Presentation Deck May 8, 2017

NVIDIA DGX SYSTEMS PURPOSE-BUILT FOR AI

Todo before next class

Artificial Intelligence Enriched User Experience with ARM Technologies

Kdb+ Transitive Comparisons

EFFICIENT INFERENCE WITH TENSORRT. Han Vanholder

TACKLING THE CHALLENGES OF NEXT GENERATION HEALTHCARE

Cortana Analytics : with Raspberry Pi and Weather Sensor

Parking Lot Practical IOT COURSE

Emergence of the Memory Centric Architectures

Raspberry Pi Tutorial 2

Shrinath Shanbhag Senior Software Engineer Microsoft Corporation

Oracle Autonomous Database

GPU ACCELERATED DATABASE MANAGEMENT SYSTEMS

IoT Market: Three Classes of Devices

Technical Manual(TM)

DGX SYSTEMS: DEEP LEARNING FROM DESK TO DATA CENTER. Markus Weber and Haiduong Vo

An Implementation of Fog Computing Attributes in an IoT Environment

NVIDIA FOR DEEP LEARNING. Bill Veenhuis

Milano. Gli atti dei convegni e più di contenuti su

How to Route Internet Traffic between A Mobile Application and IoT Device?

Transcription:

Presented @ GTC 2017 & Edge-to-Cloud Compute with MxNet AWS & NVIDIA Aran Khanna, Software Developer, AWS Miro Enev, Solutions Architect, NVIDIA 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.

Edge-To-Cloud Compute Models Overview Advantages of Hub & Spoke Case Study / Demo References & Tools

IIoT & AI AI/DL is primed to disrupt the Industrial Sector.

NVIDIA IoT

NVIDIA IoT Robots & Smart Manufacturing Generative Design Build Design Operate Inspection & Efficiency Predictive Maintenance Maintain & Repair

Amazon IoT & GreenGrass

Amazon IoT Software & NVIDIA Hardware Key Functions Data Ingest Compressed Inference Full Inference / Trained Model Query Model Training Deployment Models Cloud <-> Edge Cloud <-> Hub <-> Edge

Emerging Deployment Model Trends: Latency, Cost, Reliability, Privacy/Security Policy Cloud Elastic GPU [ Tesla ] Hub/Gateway Jetson [ Tegra ] Edge Sensors [ RasPi ]

Handoff

Use Cases Of ML at The Edge Transportation AR & VR Industrials Smart Home Robotics Body Cameras

In This Demo I will: -Show you why you should care about Machine Learning at the Edge -Walk through the value of running deep learning applications with MXNet on a set of edge devices. -Demonstrate how AWS services make managing deep learning at the edge easier than ever.

MXNet Everywhere High End: NVIDIA Jetson - 64 Bit ARM Arch64 - CUDA - Ubuntu Low End: Raspberry Pi - 32 Bit ARMv7 - ARM NEON - Raspbian

Easy as 1 2...3... 1. 2. 3. git clone https://github.com/ dmlc/mxnet.git --recursive cd mxnet && make cd python && sudo python setup.py install

ImageNet Models - Well studied - Form the base models for most image tasks - Widely available pre-trained models Try it Yourself: http://data.mxnet.io/models

Running the Model # Load the network parameters from the cloud def load_model(model_url= http://data.mxnet.io/models/imagenet/squeezenet ): urllib.urlretrieve(model_url, model.zip ) unzip(model.zip) sym, arg_params, aux_params = mx.model.load_checkpoint( Image_Model', 0) mod = mx.mod.module(symbol=sym, context= mx.cpu()) mod.bind(for_training=false,data_shapes=[('data', (1,3,224,224))]) mod.set_params(arg_params, aux_params) return mod # Predict on an Image def predict(image_path, mod): img = cv2.cvtcolor(cv2.imread(filename), cv2.color_bgr2rgb) img = cv2.resize(img, (224, 224)) img = np.swapaxes(img, 0, 2) img = np.swapaxes(img, 1, 2) img = img[np.newaxis, :] mod.forward(batch([mx.nd.array(img)])) return mod.get_outputs()[0].asnumpy()

Making This Production Ready With AWS Management Aggregation Logging - Rollout changes - Coordinate local devices - Handle network and hardware failure - Gather and centralize data from the Edge - Connect to existing systems in the cloud - Monitor performance - Analyze uptime - Diagnose failures

Enter AWS IoT and AWS Greengrass

On The Pi: SqueezeNet Object Classifier Iandola et. Al

Linking the Pi to AWS # Load the network parameters model = load_model.imagenetmodel('synset.txt', 'squeezenet_v1.1') # Run Predictions on Images Captured With the Local Camera def Callback(client, userdata, message): args = json.load(message.payload) predictions, img = model.predict_from_cam(args.model) myawsiotmqttclient.publish( sdk/test/predict, predictions,1) if(predictions[0][0] < 0.8): myawsiotmqttclient.publish( sdk/test/escalate, img, 1) mymqttclient.connect() mymqttclient.subscribe( sdk/test/control", 1, Callback)

On The TX1: ResNet-152 Object Detector Szegedy et. Al

Linking the TX1 to AWS # Load the network parameters model = load_model.imagenetmodel('synset.txt', resnet-v2') pi_model = load_model.imagenetmodel('synset.txt', 'squeezenet_v1.1') # Run Predictions on Images Captured With the Local Camera def Callback(client, userdata, message): args = json.load(message.payload) predictions = model.predict_from_img(args.img. args.model) myawsiotmqttclient.publish( sdk/test/predict, predictions, 1) # Fine tune the squeezenet model to for redeployment to Pi if(predictions[0][0] > 0.8): pi_model.update_model(args.img, predictions) mymqttclient.connect() mymqttclient.subscribe( sdk/test/escalate", 1, Callback)

Local Learning Loop Poorly Classified Picture Fine Tune Model With Accurate Classification Updated Model

Thank You! Miro Enev menev@nvidia.com Aran Khanna arankhan@amazon.com