The document discusses Google Cloud AI services including Cloud ML Engine for machine learning model training and prediction. It provides examples of using Cloud ML Engine to train models locally and in the cloud, perform distributed training, and hyperparameter tuning. It also covers deploying trained models and making predictions against them.
It’s no longer needed supercomputers and a team with PhDs from MIT to create predictive models based on data. We are witnessing innovations in machine learning that are making it an increasingly accessible field. This lecture aims to demystify machine learning through exposure to concepts and use of a number of technologies. In this talk, we will address the types of problems and the algorithms, always applied to real problems. Also, open source tools like Scikit-learn will be presented as well as a way to practice and try these ideas through competitions like Kaggle.
A brief lesson on what constitutes computational decision making, from simple regression via various classification methods to deep learning. No maths, only basic concepts to teach the lingo of machine learning to a lay audience.
Introduction To TensorFlow | Deep Learning Using TensorFlow | TensorFlow Tuto...Edureka!
In this TensorFlow tutorial, you will be learning all the basics of TensorFlow and how to create a Deep Learning Model. It includes the following topics:
1. Deep Learning vs Machine Learning
2. What is TensorFlow?
3. TensorFlow Use-Case
Daniel Shank, Data Scientist, Talla at MLconf SF 2016MLconf
Neural Turing Machines: Perils and Promise: Daniel Shank is a Senior Data Scientist at Talla, a company developing a platform for intelligent information discovery and delivery. His focus is on developing machine learning techniques to handle various business automation tasks, such as scheduling, polls, expert identification, as well as doing work on NLP. Before joining Talla as the company’s first employee in 2015, Daniel worked with TechStars Boston and did consulting work for ThriveHive, a small business focused marketing company in Boston. He studied economics at the University of Chicago.
QCon Rio - Machine Learning for EveryoneDhiana Deva
Já não são mais necessários supercomputadores e times de PhDs do MIT para a criação de modelos preditivos baseados em dados. Estamos presenciando inovações em Aprendizado de Máquina que estão tornando este campo cada vez mais acessível.
Esta palestra tem como objetivo desmistificar o aprendizado de máquina, através da exposição de conceitos e uso de uma série de tecnologias.
Serão abordados os tipos de problemas desta área(classificação, regressão, clusterização, redução de dimensionalidade, etc.), suas as etapas (normalização, treinamento, otimização, regularização, etc.) e seus algoritmos, desde regressão linear, k-means, passando por árvores de decisão e até redes neurais, sempre aplicadas a problemas reais.
Na palestra, também conheceremos ferramentas como Sckit-learn, Pandas, R, MATLAB e Amazon Machine Learning, além de uma forma para praticar e experimentar estas ideias através de competições como o Kaggle.
It’s no longer needed supercomputers and a team with PhDs from MIT to create predictive models based on data. We are witnessing innovations in machine learning that are making it an increasingly accessible field. This lecture aims to demystify machine learning through exposure to concepts and use of a number of technologies. In this talk, we will address the types of problems and the algorithms, always applied to real problems. Also, open source tools like Scikit-learn will be presented as well as a way to practice and try these ideas through competitions like Kaggle.
A brief lesson on what constitutes computational decision making, from simple regression via various classification methods to deep learning. No maths, only basic concepts to teach the lingo of machine learning to a lay audience.
Introduction To TensorFlow | Deep Learning Using TensorFlow | TensorFlow Tuto...Edureka!
In this TensorFlow tutorial, you will be learning all the basics of TensorFlow and how to create a Deep Learning Model. It includes the following topics:
1. Deep Learning vs Machine Learning
2. What is TensorFlow?
3. TensorFlow Use-Case
Daniel Shank, Data Scientist, Talla at MLconf SF 2016MLconf
Neural Turing Machines: Perils and Promise: Daniel Shank is a Senior Data Scientist at Talla, a company developing a platform for intelligent information discovery and delivery. His focus is on developing machine learning techniques to handle various business automation tasks, such as scheduling, polls, expert identification, as well as doing work on NLP. Before joining Talla as the company’s first employee in 2015, Daniel worked with TechStars Boston and did consulting work for ThriveHive, a small business focused marketing company in Boston. He studied economics at the University of Chicago.
QCon Rio - Machine Learning for EveryoneDhiana Deva
Já não são mais necessários supercomputadores e times de PhDs do MIT para a criação de modelos preditivos baseados em dados. Estamos presenciando inovações em Aprendizado de Máquina que estão tornando este campo cada vez mais acessível.
Esta palestra tem como objetivo desmistificar o aprendizado de máquina, através da exposição de conceitos e uso de uma série de tecnologias.
Serão abordados os tipos de problemas desta área(classificação, regressão, clusterização, redução de dimensionalidade, etc.), suas as etapas (normalização, treinamento, otimização, regularização, etc.) e seus algoritmos, desde regressão linear, k-means, passando por árvores de decisão e até redes neurais, sempre aplicadas a problemas reais.
Na palestra, também conheceremos ferramentas como Sckit-learn, Pandas, R, MATLAB e Amazon Machine Learning, além de uma forma para praticar e experimentar estas ideias através de competições como o Kaggle.
Slides from the presentation given at M^3 conference: http://www.mcubed.london/
The idea is to use 3 statements to describe and start to work with the TensorFlow library.
Hussein Mehanna, Engineering Director, ML Core - Facebook at MLconf ATL 2016MLconf
Applying Deep Learning at Facebook Scale: Facebook leverages Deep Learning for various applications including event prediction, machine translation, natural language understanding and computer vision at a very large scale. There are more than a billion users logging on to Facebook every daily generating thousands of posts per second and uploading more than a billion images and videos every day. This talk will explain how Facebook scaled Deep Learning inference for realtime applications with latency budgets in the milliseconds.
Data Science, Machine Learning and Neural NetworksBICA Labs
Lecture briefly overviewing state of the art of Data Science, Machine Learning and Neural Networks. Covers main Artificial Intelligence technologies, Data Science algorithms, Neural network architectures and cloud computing facilities enabling the whole stack.
Language translation with Deep Learning (RNN) with TensorFlowS N
The author is going to take you into the realm of Recurrent Neural Network (RNN). He will be training a sequence to sequence model on a dataset of English and French sentences that can translate new (unseen) sentences from English to French.
This will be a walkthrough of an end to end technique to train a Deep RNN model. You will learn to build various components necessary to build a Sequence-to-Sequence model.
You will learn about the fundamentals of Deep Learning, mainly RNN, concepts that will be required in this solution. A familiarity of Deep Learning concepts would be handy, but most of the concepts used in this example will be covered during the demo.
Technologies to be used:
Python, Jupyter, TensorFlow, FloydHub
Source code: https://github.com/syednasar/deeplearning/blob/master/language-translation/dlnd_language_translation.ipynb
...
Practical deep learning for computer visionEran Shlomo
This is the presentation given in TLV DLD 2017. In this presentation we walk through the planning and implemintation of deeplearning solution for image recognition, with focus on the data.
It is based on the work we do at dataloop.ai and its customers.
This is a single day course, allows the learner to get experience with the basic details of deep learning, first half is building a network using python/numpy only and the second half we build the more advanced netwrok using TensorFlow/Keras.
At the end you will find a list of usefull pointers to continue.
course git: https://gitlab.com/eshlomo/EazyDnn
In this deck, Huihuo Zheng from Argonne National Laboratory presents: Data Parallel Deep Learning.
"The Argonne Training Program on Extreme-Scale Computing (ATPESC) provides intensive, two weeks of training on the key skills, approaches, and tools to design, implement, and execute computational science and engineering applications on current high-end computing systems and the leadership-class computing systems of the future."
Watch the video: https://wp.me/p3RLHQ-lsl
Learn more: https://extremecomputingtraining.anl.gov/archive/atpesc-2019/agenda-2019/
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
A simplified way of approaching machine learning and deep learning from the ground up. The case for deep learning and an attempt to develop intuition for how/why it works. Advantages, state-of-the-art, and trends.
Presented at NYU Center for Genomics for NY Deep Learning Meetup
Deep Learning with TensorFlow: Understanding Tensors, Computations Graphs, Im...Altoros
1. The elements of Neural Networks: Weights, Biases, and Gating functions
2. MNIST (Hand writing recognition) using simple NN in TensorFlow (Introduce Tensors, Computation Graphs)
3. MNIST using Convolution NN in TensorFlow
4. Understanding words and sentences as Vectors
5. word2vec in TensorFlow
Video: http://videos.re-work.co/videos/464-agile-deep-learning
Deep Learning has been called the ‘new electricity’ — transforming every industry. Innovative architectures and applications receive deserved attention. But to turn innovation into value requires integrating deep learning into practical technology products. Such products, including Spotify's, are often developed following the principles of agile. This talk focuses on approaching deep learning in an agile way and on integrating deep learning into the agile cadence of a modern software development organization.
Brief presentation about keras framework. The propose of this presentation is to give some ideas about how it works and its main functionalities. In addition, is also shown a function to create different models from a config file.
For the full video of this presentation, please visit:
https://www.edge-ai-vision.com/2021/02/practical-guide-to-implementing-deep-neural-network-inferencing-at-the-edge-a-presentation-from-zebra-technologies/
Toly Kotlarsky, Distinguished Member of the Technical Staff in R&D at Zebra Technologies, presents the “Practical Guide to Implementing Deep Neural Network Inferencing at the Edge” tutorial at the September 2020 Embedded Vision Summit.
In this presentation, Kotlarsky explores practical aspects of implementing a pre-trained deep neural network (DNN) inference on typical edge processors. First, he briefly touches on how we evaluate the accuracy of DNNs for use in real-world applications. Next, he explains the process for converting a trained model in TensorFlow into formats suitable for deployment at the edge and examines a simple, generic C++ real-time inference application that can be deployed on a variety of hardware platforms
Kotlarsky then outlines a method for evaluating the performance of edge DNN implementations and shows the results of utilizing this method to benchmark the performance of three popular edge computing platforms: The Google Coral (based on the Edge TPU), NVIDIA Jetson Nano and Raspberry Pi 3.
Nikhil Garg, Engineering Manager, Quora at MLconf SF 2016MLconf
Building a Machine Learning Platform at Quora: Each month, over 100 million people use Quora to share and grow their knowledge. Machine learning has played a critical role in enabling us to grow to this scale, with applications ranging from understanding content quality to identifying users’ interests and expertise. By investing in a reusable, extensible machine learning platform, our small team of ML engineers has been able to productionize dozens of different models and algorithms that power many features across Quora.
In this talk, I’ll discuss the core ideas behind our ML platform, as well as some of the specific systems, tools, and abstractions that have enabled us to scale our approach to machine learning.
Image Classification Done Simply using Keras and TensorFlow Rajiv Shah
This presentation walks through the process of building an image classifier using Keras with a TensorFlow backend. It will give a basic understanding of image classification and show the techniques used in industry to build image classifiers. The presentation will start with building a simple convolutional network, augmenting the data, using a pretrained network, and finally using transfer learning by modifying the last few layers of a pretrained network. The classification will be based on the classic example of classifying cats and dogs. The code for the presentation can be found at https://github.com/rajshah4/image_keras, and the presentation will discuss how to extend the code to your own pictures to make a custom image classifier.
Anomaly Detection in Time-Series Data using the Elastic Stack by Henry PakData Con LA
Abstract:- The Elastic has released a commercial machine learning plugin that allows you to create a model of your time series data using an unsupervised machine learning approach. Walk through a few common use cases to see how this plugin may help with finding anomalies in your data.
Slides from the presentation given at M^3 conference: http://www.mcubed.london/
The idea is to use 3 statements to describe and start to work with the TensorFlow library.
Hussein Mehanna, Engineering Director, ML Core - Facebook at MLconf ATL 2016MLconf
Applying Deep Learning at Facebook Scale: Facebook leverages Deep Learning for various applications including event prediction, machine translation, natural language understanding and computer vision at a very large scale. There are more than a billion users logging on to Facebook every daily generating thousands of posts per second and uploading more than a billion images and videos every day. This talk will explain how Facebook scaled Deep Learning inference for realtime applications with latency budgets in the milliseconds.
Data Science, Machine Learning and Neural NetworksBICA Labs
Lecture briefly overviewing state of the art of Data Science, Machine Learning and Neural Networks. Covers main Artificial Intelligence technologies, Data Science algorithms, Neural network architectures and cloud computing facilities enabling the whole stack.
Language translation with Deep Learning (RNN) with TensorFlowS N
The author is going to take you into the realm of Recurrent Neural Network (RNN). He will be training a sequence to sequence model on a dataset of English and French sentences that can translate new (unseen) sentences from English to French.
This will be a walkthrough of an end to end technique to train a Deep RNN model. You will learn to build various components necessary to build a Sequence-to-Sequence model.
You will learn about the fundamentals of Deep Learning, mainly RNN, concepts that will be required in this solution. A familiarity of Deep Learning concepts would be handy, but most of the concepts used in this example will be covered during the demo.
Technologies to be used:
Python, Jupyter, TensorFlow, FloydHub
Source code: https://github.com/syednasar/deeplearning/blob/master/language-translation/dlnd_language_translation.ipynb
...
Practical deep learning for computer visionEran Shlomo
This is the presentation given in TLV DLD 2017. In this presentation we walk through the planning and implemintation of deeplearning solution for image recognition, with focus on the data.
It is based on the work we do at dataloop.ai and its customers.
This is a single day course, allows the learner to get experience with the basic details of deep learning, first half is building a network using python/numpy only and the second half we build the more advanced netwrok using TensorFlow/Keras.
At the end you will find a list of usefull pointers to continue.
course git: https://gitlab.com/eshlomo/EazyDnn
In this deck, Huihuo Zheng from Argonne National Laboratory presents: Data Parallel Deep Learning.
"The Argonne Training Program on Extreme-Scale Computing (ATPESC) provides intensive, two weeks of training on the key skills, approaches, and tools to design, implement, and execute computational science and engineering applications on current high-end computing systems and the leadership-class computing systems of the future."
Watch the video: https://wp.me/p3RLHQ-lsl
Learn more: https://extremecomputingtraining.anl.gov/archive/atpesc-2019/agenda-2019/
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
A simplified way of approaching machine learning and deep learning from the ground up. The case for deep learning and an attempt to develop intuition for how/why it works. Advantages, state-of-the-art, and trends.
Presented at NYU Center for Genomics for NY Deep Learning Meetup
Deep Learning with TensorFlow: Understanding Tensors, Computations Graphs, Im...Altoros
1. The elements of Neural Networks: Weights, Biases, and Gating functions
2. MNIST (Hand writing recognition) using simple NN in TensorFlow (Introduce Tensors, Computation Graphs)
3. MNIST using Convolution NN in TensorFlow
4. Understanding words and sentences as Vectors
5. word2vec in TensorFlow
Video: http://videos.re-work.co/videos/464-agile-deep-learning
Deep Learning has been called the ‘new electricity’ — transforming every industry. Innovative architectures and applications receive deserved attention. But to turn innovation into value requires integrating deep learning into practical technology products. Such products, including Spotify's, are often developed following the principles of agile. This talk focuses on approaching deep learning in an agile way and on integrating deep learning into the agile cadence of a modern software development organization.
Brief presentation about keras framework. The propose of this presentation is to give some ideas about how it works and its main functionalities. In addition, is also shown a function to create different models from a config file.
For the full video of this presentation, please visit:
https://www.edge-ai-vision.com/2021/02/practical-guide-to-implementing-deep-neural-network-inferencing-at-the-edge-a-presentation-from-zebra-technologies/
Toly Kotlarsky, Distinguished Member of the Technical Staff in R&D at Zebra Technologies, presents the “Practical Guide to Implementing Deep Neural Network Inferencing at the Edge” tutorial at the September 2020 Embedded Vision Summit.
In this presentation, Kotlarsky explores practical aspects of implementing a pre-trained deep neural network (DNN) inference on typical edge processors. First, he briefly touches on how we evaluate the accuracy of DNNs for use in real-world applications. Next, he explains the process for converting a trained model in TensorFlow into formats suitable for deployment at the edge and examines a simple, generic C++ real-time inference application that can be deployed on a variety of hardware platforms
Kotlarsky then outlines a method for evaluating the performance of edge DNN implementations and shows the results of utilizing this method to benchmark the performance of three popular edge computing platforms: The Google Coral (based on the Edge TPU), NVIDIA Jetson Nano and Raspberry Pi 3.
Nikhil Garg, Engineering Manager, Quora at MLconf SF 2016MLconf
Building a Machine Learning Platform at Quora: Each month, over 100 million people use Quora to share and grow their knowledge. Machine learning has played a critical role in enabling us to grow to this scale, with applications ranging from understanding content quality to identifying users’ interests and expertise. By investing in a reusable, extensible machine learning platform, our small team of ML engineers has been able to productionize dozens of different models and algorithms that power many features across Quora.
In this talk, I’ll discuss the core ideas behind our ML platform, as well as some of the specific systems, tools, and abstractions that have enabled us to scale our approach to machine learning.
Image Classification Done Simply using Keras and TensorFlow Rajiv Shah
This presentation walks through the process of building an image classifier using Keras with a TensorFlow backend. It will give a basic understanding of image classification and show the techniques used in industry to build image classifiers. The presentation will start with building a simple convolutional network, augmenting the data, using a pretrained network, and finally using transfer learning by modifying the last few layers of a pretrained network. The classification will be based on the classic example of classifying cats and dogs. The code for the presentation can be found at https://github.com/rajshah4/image_keras, and the presentation will discuss how to extend the code to your own pictures to make a custom image classifier.
Anomaly Detection in Time-Series Data using the Elastic Stack by Henry PakData Con LA
Abstract:- The Elastic has released a commercial machine learning plugin that allows you to create a model of your time series data using an unsupervised machine learning approach. Walk through a few common use cases to see how this plugin may help with finding anomalies in your data.
Dat de snelle ontwikkelingen op het gebied van digitale technologie de samenleving veranderen is overduidelijk. We beseffen alleen nog niet altijd hoe ingrijpend die veranderingen zijn. Het gaat niet enkel over automatisering, of over het efficiënter maken van wat we al deden door het inzetten van nieuwe technologie. Steeds meer merk je dat de digitale revolutie de regels van het spel zelf verandert. Neem Uber als voorbeeld: het grootste taxibedrijf ter wereld bezit zelf geen enkele auto en zet zo het klassieke model van zaken doen helemaal op zijn kop.
Ook in de sfeer van educatie, gemeenschapsvorming, maatschappelijke en culturele actie zie je gelijkaardige ontwikkelingen.
In een dergelijke context verandert de manier waarop we aan sociaal-cultureel werk doen fundamenteel. De digitale transformatie dwingt je om de rol die je als organisatie speelt en de manier waarop je je doelen nastreeft grondig te bevragen en voor een stuk te herdefiniëren.
Hoe kunnen we digitale transformatie echt begrijpen? Wat betekenen deze ontwikkelingen voor het sociaal-cultureel werk? En hoe kunnen we ons ertoe verhouden? Jo Caudron, co-auteur van het boek 'Digital transformation', legt aan de hand van 7 metaforen uit wat digitale transformatie precies inhoudt.
Presentatie op studiedag "Sociaal-cultureel werk in digitale transformatie" (9 juni 2016).
NUS-ISS Learning Day 2016 - Big Data AnalyticsNUS-ISS
A real-time descriptive data analytics of your data seating inside of your NoSQL database. A time series data will be index to the lucene-based search server called ElasticSearch. This indexed data will then be visualised through the visualisation tool called Kibana. This tool can show charts, trends, maps and graphs based on your data. You can customise the filters to really get what you want from your data. Learn how you can quickly understand and get insights from their data.
Mainframe Customer Education Webcast: New Ironstream Facilities for Enhanced ...Precisely
Check out our latest Mainframe Customer Education Webcast, featuring new Ironstream facilities for enhanced z/OS Analytics. Product Management Directors Ed Wrazen and Ed Hallock spoke about what’s new in the Ironstream z/OS data forwarder, as well as new features and facilities including:
• Data Loss Protection
• Advanced Filtering for SMF data
• Splunk Applications for Ironstream
You'll also learn about integration with Splunk’s IT Service Intelligence for monitoring the availability of critical business services running on z/OS platforms.
A fast-paced introduction to Deep Learning concepts, such as activation functions, cost functions, backpropagation, and then a quick dive into CNNs. Basic knowledge of vectors, matrices, and elementary calculus (derivatives), are helpful in order to derive the maximum benefit from this session.
Next we'll see a simple neural network using Keras, followed by an introduction to TensorFlow and TensorBoard. (Bonus points if you know Zorn's Lemma, the Well-Ordering Theorem, and the Axiom of Choice.)
A Tale of Three Deep Learning Frameworks: TensorFlow, Keras, & PyTorch with B...Databricks
We all know what they say – the bigger the data, the better. But when the data gets really big, how do you mine it and what deep learning framework to use? This talk will survey, with a developer’s perspective, three of the most popular deep learning frameworks—TensorFlow, Keras, and PyTorch—as well as when to use their distributed implementations.
We’ll compare code samples from each framework and discuss their integration with distributed computing engines such as Apache Spark (which can handle massive amounts of data) as well as help you answer questions such as:
As a developer how do I pick the right deep learning framework?
Do I want to develop my own model or should I employ an existing one?
How do I strike a trade-off between productivity and control through low-level APIs?
What language should I choose?
In this session, we will explore how to build a deep learning application with Tensorflow, Keras, or PyTorch in under 30 minutes. After this session, you will walk away with the confidence to evaluate which framework is best for you.
This presentation focuses on Deep Learning (DL) concepts, such as neural networks, backprop, activation functions, and Convolutional Neural Networks. You'll also learn how to incorporate Deep Learning in Android applications. Basic knowledge of matrices is helpful for this session, which is targeted primarily to beginners.
Kaz Sato, Evangelist, Google at MLconf ATL 2016MLconf
Machine Intelligence at Google Scale: Tensor Flow and Cloud Machine Learning: The biggest challenge of Deep Learning technology is the scalability. As long as using single GPU server, you have to wait for hours or days to get the result of your work. This doesn’t scale for production service, so you need a Distributed Training on the cloud eventually. Google has been building infrastructure for training the large scale neural network on the cloud for years, and now started to share the technology with external developers. In this session, we will introduce new pre-trained ML services such as Cloud Vision API and Speech API that works without any training. Also, we will look how TensorFlow and Cloud Machine Learning will accelerate custom model training for 10x – 40x with Google’s distributed training infrastructure.
This presentation introduces Deep Learning (DL) concepts, such as neural neworks, backprop, activation functions, and Convolutional Neural Networks, followed by an Angular application that uses TypeScript in order to replicate the Tensorflow playground.
This talk was presented in Startup Master Class 2017 - http://aaiitkblr.org/smc/ 2017 @ Christ College Bangalore. Hosted by IIT Kanpur Alumni Association and co-presented by IIT KGP Alumni Association, IITACB, PanIIT, IIMA and IIMB alumni.
My co-presenter was Biswa Gourav Singh. And contributor was Navin Manaswi.
http://dataconomy.com/2017/04/history-neural-networks/ - timeline for neural networks
Machine learning with Go. There are some libraries like gonum and gorgonia that allows you to do machine learning. You can also train your deep learning models with python and TensorFlow and use the TensorFlow libraries to read exported models. Go-ONNX is to come.
Centric - Jaap huisprijzen, GTST, The Bold, IKEA en IENS. Zomaar wat toepassi...BigDataExpo
Tijdens deze presentatie wordt duidelijk hoe je machine learning kunt toepassen in het dagelijks leven. Denk aan het kopen van een huis, het kijken van Goede Tijden Slechte Tijden, shoppen bij IKEA en het bezoeken van restaurants.
In this session we'll dive into the journey that Google chooses to take in order focus on AI: what was the mindset, what were the challenges and what is the direction for the future.
Pacmed - Machine Learning in health care: opportunities and challanges in pra...BigDataExpo
The potential of personalized medicine based on machine learning is huge, but big challenges must be overcome to implement this technology in practice. Hidde will discuss both sides of the story, including a case study on the intensive care.
De Toekomst Verkenner is een ‘award winning’ innovatie van PGGM, die in een rap temp doorontwikkeling naar een platform maakt.
In zijn presentatie zal Mladen Sančanin vertellen hoe PGGM real time data en algoritmes heeft ingezet om dit platform te bouwen en hoe PGGM innovaties vanuit haar ‘Big Data Lab’ ondersteunt?
In een half uur worden veel ervaringen gedeeld over het opzetten van innovatieprojecten gebruik makend van data en het inrichten van data lab in een corporate omgeving.
Universiteit Utrecht & gghdc - Wat zijn de gezondheidseffecten van omgeving e...BigDataExpo
Het GGHDC onderzoekt wat de gezondheidseffecten zijn van omgeving en leefstijl in relatie tot het dagelijks leven van mensen. Het onderzoekscentrum is opgebouwd rond een gedeelde data- infrastructuur van de Universiteit Utrecht en het Universitair Medisch Centrum Utrecht (UMCU).
Rob van Kranenburg - Kunnen we ons een sociaal krediet systeem zoals in het o...BigDataExpo
IoT, Big Data, AI creëren een nieuwe situatie met betrekking tot het nemen van beslissingen door beleidsmakers. Toch verschuift er weinig in ons democratisch bestel, terwijl onze data in handen zijn van GAFA, China en andere nieuwe vormen van bestuur die nog ontstaan in de digitale transitie. Wij, in Europa, staan stil.
OrangeNXT - High accuracy mapping from videos for efficient fiber optic cable...BigDataExpo
Construction companies such as BAM Infra Telecom rely on accurate, up-to-date maps. Google Maps isn’t enough, but doing on-site surveys is expensive and time-consuming. However, driving through and recording 360° video from a car is cheap and easy. Using machine learning, we turn videos into highly accurate maps.
Dynniq & GoDataDriven - Shaping the future of traffic with IoT and AIBigDataExpo
Dynniq is a high-tech, innovative company offering smart mobility solutions and services internationally. We will present advanced IoT use cases Dynniq is working on, and share how GoDataDriven helps set up an AI capability. We will share our learnings, and show what makes data science in the mobility domain unique.
Teleperformance - Smart personalized service door het gebruik van Data Science BigDataExpo
Bij Teleperformance helpen we klanten waarde toe te voegen aan het klanttraject. We gebruiken Data Science voor onze Omnichannel-klantinteracties om de behoeften van de klant te voorspellen, zodat we het beste antwoord kunnen geven.
FunXtion - Interactive Digital Fitness with Data AnalyticsBigDataExpo
Digital is the new Personal. FunXtion Interactive is een interactieve trainingservaring voor zowel binnen als buiten de sportschool. FunXtion is revolutionair in de fitness branche en volledig data driven, by design. FunXtion laat zien hoe zij real-time data gebruiken voor ondersteuning van beslissingen, proces automatisering, personalisatie en product innovatie.
fashionTrade - Vroeger noemde we dat Big DataBigDataExpo
Big Data was de verzamelnaam voor alles wat je nog niet deed, maar al wel door Google of Amazon was uitgevonden. Inmiddels doen we al die dingen wel dus heet productaanbevelingen weer gewoon productaanbevelingen, fraudebestrijding weer fraudebestrijding, en spraakherkenning nog steeds spraakherkenning; geen Big Data. Geeft niet, want nu is er AI. Deze keynote legt uit of dat anders is, en waarom.
BigData Republic - Industrializing data science: a view from the trenchesBigDataExpo
What does it take to bring machine learning algorithms to production and start delivering business value? How can teams of data scientists and engineers effectively collaborate on a single product, integrate with existing IT systems and keep business stakeholders involved? Using real-life examples, we discuss the challenges and best practices.
Bicos - Hear how a top sportswear company produced cutting-edge data infrastr...BigDataExpo
Industry expert Dave Vanhoudt will set out his vision for the future of data infrastructure. Dave will highlight the key role automation must play in any data infrastructure strategy today, drawing on his current role with Medtronic, and past experiences at AB Inbev, Baxter, BMW and Nike.
Endrse - Next level online samenwerkingen tussen personalities en merken met ...BigDataExpo
Digitaal is vrijwel alles meetbaar. Maar het is vaak een uitdaging om de impact van samenwerkingen tussen influencers (topsporters) en bedrijven te analyseren. Start-up Endrse gebruikt AI om socialmediacontent te analyseren om content van influencers en bedrijven beter op elkaar te laten aansluiten. Zo maak je impact bij het publiek!
Bovag - Refine-IT - Proces optimalisatie in de automotive sectorBigDataExpo
De ontwikkelingen in de automotive sector gaan snel: elektrisch rijdende auto’s, de snelle groei van private lease, over the air connectiviteit, services on the demand en advanced driver assistance is zo maar een greep uit deze ontwikkelingen. Voorbeelden van (big) data ontwikkelingen die van grote invloed zijn op de automotive retail. De transitie naar een nieuw verdienmodel daagt uit tot samenwerken en datagedreven procesoptimalisatie.
Wilco Schellevis, directeur van Refine-IT en Renate Weggemans, manager strategie en beleid, bij BOVAG Autodealers, nemen u mee in de case Dely-App. Een mooi staaltje samenwerken en datagedreven procesoptimalisatie in de automotive retail; gevangen in één app.
Schiphol - Optimale doorstroom van passagiers op Schiphol dankzij slimme data...BigDataExpo
Schiphol is Europa’s best connected airport en verwerkt op piekdagen tot 235.000 passagiers. Om deze soepel door de processen te leiden is een betrouwbare prognose van de drukte noodzakelijk. Schiphol laat zien hoe zij datatoepassingen ontwikkelt om het aantal reizigers zo accuraat mogelijk te voorspellen en hiermee processen in te richten.
Veco - Big Data in de Supply Chain: Hoe Process Mining kan helpen kosten te r...BigDataExpo
Veco is marktleider op het gebied van het ontwerpen en vervaardigen van precisie delen middels electroformeren. In deze presentatie zal uitgeleverd worden hoe Veco succesvol Process Mining heeft ingezet in de productie om doorlooptijd te reduceren en new business te creëren. Tevens wordt uitgelegd wat Process Mining is.
Rabobank - There is something about DataBigDataExpo
Technologische mogelijkheden en GDPR, een continue clash? En hoe staat het met de het ethisch (her)gebruik van data? Leer in deze sessie van Rabobank’s Big Data journey en krijg inzicht in: organisatorische keuzes, data Lab technologie visie & data strategie, als enabler en accelerator van digitale innovatie en transformatie.
VU Amsterdam - Big data en datagedreven waardecreatie: valt er nog iets te ki...BigDataExpo
In zijn presentatie gaat Frans Feldberg in op het ‘Waarom, Wat, en Hoe’ van big data en datagedreven business model innovation. Hoe is de wereld, als het om data gaat, de laatste jaren veranderd? Waarom zijn big data, business analytics en kunstmatige intelligentie belangrijke digitale innovaties die hoog op menig managementagenda staat en waarom investeren organisaties aanzienlijk in big data en data science? Hoe kunnen organisaties waarde met data creëren door zowel het verbeteren van het bestaande business model als door nieuwe data-gedreven business modellen te ontwikkelen. Dit zijn vragen die in zijn presentatie beantwoord zullen worden.
Booking.com - Data science and experimentation at Booking.com: a data-driven ...BigDataExpo
At Booking.com we have experienced what a data driven organisation means for creating business impact. And what looks it like, when experimentation is part of your company culture.
During this session we will share our experiences and learnings on how data science and experimentation go hand in go.
Techniques to optimize the pagerank algorithm usually fall in two categories. One is to try reducing the work per iteration, and the other is to try reducing the number of iterations. These goals are often at odds with one another. Skipping computation on vertices which have already converged has the potential to save iteration time. Skipping in-identical vertices, with the same in-links, helps reduce duplicate computations and thus could help reduce iteration time. Road networks often have chains which can be short-circuited before pagerank computation to improve performance. Final ranks of chain nodes can be easily calculated. This could reduce both the iteration time, and the number of iterations. If a graph has no dangling nodes, pagerank of each strongly connected component can be computed in topological order. This could help reduce the iteration time, no. of iterations, and also enable multi-iteration concurrency in pagerank computation. The combination of all of the above methods is the STICD algorithm. [sticd] For dynamic graphs, unchanged components whose ranks are unaffected can be skipped altogether.
1. a 30 min short walk
Robert Saxby - Big Data Product Specialist
2.
3. Sustainability
Google datacenters have half the
overhead of typical industry data centers
Largest private investor in renewables: $2
billion generating 3.2 GW
Applying Machine Learning produced
40% reduction in cooling energy
5. App DeveloperData Scientist
Build custom modelsUse/extend OSS SDK Use pre-built models
ML researcher
Cloud MLE ML Perception services
End to End: Google Cloud AI Spectrum
6. App DeveloperData Scientist
Build custom modelsUse/extend OSS SDK Use pre-built models
ML researcher
Cloud MLE ML Perception services
End to End: Google Cloud AI Spectrum
7. Proprietary + Confidential
What is TensorFlow?
● A system for distributed, parallel machine learning
● It’s based on general-purpose dataflow graphs
● It targets heterogeneous devices
○ A single PC with CPU
○ A single PC with GPU(s)
○ A mobile device
○ Clusters of 100s or 1000s of CPUs, GPUs and TPUs
8. Proprietary + Confidential
Another data flow system
MatMul
Add Relu
biases
weights
examples
labels
Xent
Graph of Nodes, also called Operations or ops
9. Proprietary + Confidential
With tensors
MatMul
Add Relu
biases
weights
examples
labels
Xent
Edges are N-dimensional arrays: Tensors
10. Proprietary + Confidential
What’s in a name?
0 Scalar (magnitude only) s = 483
1 Vector (magnitude and direction) v = [1.1, 2.2, 3.3]
2 Matrix (table of numbers) m = [[1, 2, 3], [4, 5, 6], [7, 8, 9]]
3 3-Tensor (cube of numbers) t = [[[2], [4], [6]], [[8], [10], [12]], [[14], [16], [18]]]
4 n-Tensor (you get the idea) ....
14. TensorFlow Distributed Execution Engine
CPU GPU Android iOS ...
C++ FrontendPython Frontend ...
Layers
Estimator
Models in a box
Train and evaluate
models
Build models
Keras
Model
Canned Estimators
15. Proprietary + Confidential
Artificial Intelligence
The science of making things smart
Neural Network
A type of algorithm in machine learning
Machine Learning
Building machines that can learn
16. Proprietary + Confidential
The popular imagination of what ML is
Lots of data Magical resultsComplex mathematics in multidimensional spaces
17. Proprietary + Confidential
In reality, ML is
Collect
data
Create the
model
Refine the
model
Understand
and prepare
the data
Serve the
model
Define
objectives
18. Proprietary + Confidential
In reality, ML is
Collect
data
Create the
model
Refine the
model
Understand
and prepare
the data
Serve the
model
Define
objectives
26. Proprietary + Confidential
Predictions Images Weights Biases
Y[100, 10] X[100, 784] W[784,10] b[10]
matrix multiply
broadcast
on all lines
applied line
by line
tensor shapes in [ ]
Softmax on a batch of images
30. Proprietary + Confidential
import tensorflow as tf
X = tf.placeholder(tf.float32, [None, 28, 28, 1])
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))
init = tf.initialize_all_variables()
# model
Y=tf.nn.softmax(tf.matmul(tf.reshape(X,[-1, 784]), W) + b)
# placeholder for correct answers
Y_ = tf.placeholder(tf.float32, [None, 10])
# loss function
cross_entropy = -tf.reduce_sum(Y_ * tf.log(Y))
# % of correct answers found in batch
is_correct = tf.equal(tf.argmax(Y,1), tf.argmax(Y_,1))
accuracy = tf.reduce_mean(tf.cast(is_correct,tf.float32))
optimizer = tf.train.GradientDescentOptimizer(0.003)
train_step = optimizer.minimize(cross_entropy)
sess = tf.Session()
sess.run(init)
for i in range(10000):
# load batch of images and correct answers
batch_X, batch_Y = mnist.train.next_batch(100)
train_data={X: batch_X, Y_: batch_Y}
# train
sess.run(train_step, feed_dict=train_data)
# success ? add code to print it
a,c = sess.run([accuracy, cross_entropy], feed=train_data)
# success on test data ?
test_data={X:mnist.test.images, Y_:mnist.test.labels}
a,c = sess.run([accuracy, cross_entropy], feed=test_data)
initialisation
model
success metrics
training step
Run
The whole code
31. Workshop
Self-paced code lab (summary below ↓): goo.gl/mVZloU
Code: github.com/martin-gorner/tensorflow-mnist-tutorial
1-5. Theory (install then sit back and listen or read)
Neural networks 101: softmax, cross-entropy,
mini-batching, gradient descent, hidden layers, sigmoids,
and how to implement them in Tensorflow
6. Practice (full instructions for this step)
Open file: mnist_1.0_softmax.py
Run it, play with the visualisations (keyboard shortcuts
on previous slide), read and understand the code as well
as the basic structure of a Tensorflow program.
7. Practice (full instructions for this step)
Start from the file mnist_1.0_softmax.py and add one
or two hidden layers.
Solution in: mnist_2.0_five_layers_sigmoid.py
8. Practice (full instructions for this step)
Special care for deep neural networks: use RELU
activation functions, use a better optimiser, initialise
weights with random values and beware of the log(0)
9-10. Practice (full instructions for this step)
Use a decaying learning rate and then add dropout
Solution in: mnist_2.2_five_layers_relu_lrdecay_dropout.py
11. Theory (sit back and listen or read)
Convolutional networks
12. Practice (full instructions for this step)
Replace your model with a convolutional network,
without dropout.
Solution in: mnist_3.0_convolutional.py
13. Challenge (full instructions for this step)
Try a bigger neural network (good hyperparameters on
slide 43) and add dropout on the last layer to get >99%
Solution in: mnist_3.0_convolutional_bigger_dropout.py
?
?
33. Proprietary + Confidential
Machine Learning on any data, of any size
Cloud ML Engine
Portable models with TensorFlow
Services are designed to work together
Managed distributed training infrastructure
that supports CPUs and GPUs
Automatic hyperparameter tuning
34. Custom Estimators: The Model
https://github.com/GoogleCloudPlatform/cloudml-samples/tree/master/census
...
def _model_fn(mode, features, labels):
...
if mode == Modes.PREDICT:
...
return tf.estimator.EstimatorSpec(mode, predictions=predictions, export_outputs=export_outputs)
...
if mode == Modes.TRAIN:
...
return tf.estimator.EstimatorSpec(mode, loss=loss, train_op=train_op)
...
35. https://github.com/GoogleCloudPlatform/cloudml-samples/tree/master/census
Custom Estimators: The Task
...
train_input = lambda: model.generate_input_fn(hparams.train_files, num_epochs=hparams.num_epochs,
batch_size=hparams.train_batch_size)
...
"""This function is used by learn_runner to create an Experiment which executes model code
provided in the form of an Estimator and input functions."""
def _experiment_fn(run_config, hparams):
tf.estimator.Estimator(
model.generate_model_fn(
...
),
train_input_fn=train_input,
eval_input_fn=eval_input,
**experiment_args
)
...
36. Proprietary + Confidential
Running locally
gcloud ml-engine local train
--module-name trainer.task --package-path trainer/
--
--train-files $TRAIN_DATA --eval-files $EVAL_DATA --train-steps 1000 --job-dir $MODEL_DIR
training
data
evaluation
data
output
directory
train locally
37. Proprietary + Confidential
Single trainer running in the cloud
gcloud ml-engine jobs submit training $JOB_NAME --job-dir $OUTPUT_PATH
--runtime-version 1.0 --module-name trainer.task --package-path trainer/ --region $REGION
--
--train-files $TRAIN_DATA --eval-files $EVAL_DATA --train-steps 1000 --verbosity DEBUG
train in the cloud
region
Google cloud storage
location
38. Proprietary + Confidential
Distributed training in the cloud
gcloud ml-engine jobs submit training $JOB_NAME --job-dir $OUTPUT_PATH
--runtime-version 1.0 --module-name trainer.task --package-path trainer/ --region $REGION
--scale-tier STANDARD_1
--
--train-files $TRAIN_DATA --eval-files $EVAL_DATA --train-steps 1000 --verbosity DEBUG
distributed
39. Proprietary + Confidential
In reality, ML is
Collect
data
Create the
model
Refine the
model
Understand
and prepare
the data
Serve the
model
Define
objectives
41. Proprietary + Confidential
Hyperparameter tuning
● Automatic hyperparameter tuning service
● Build better performing models faster and save
many hours of manual tuning
● Google-developed search (Bayesian Optimisation)
algorithm efficiently finds better hyperparameters
for your model/dataset
HyperParam #1
Objective
We want to find this
Not these
https://cloud.google.com/blog/big-data/2017/08/hyperparameter-tuning-in-cloud-machine-learning-engine-using-bayesian-optimization
43. Proprietary + Confidential
Hyperparameter tuning
trainingInput:
hyperparameters:
goal: MAXIMIZE
hyperparameterMetricTag: accuracy
maxTrials: 4
maxParallelTrials: 2
params:
- parameterName: first-layer-size
type: INTEGER
minValue: 50
maxValue: 500
scaleType: UNIT_LINEAR_SCALE
...
...
# Construct layers sizes with exponetial decay
hidden_units=[
max(2, int(hparams.first_layer_size *
hparams.scale_factor**i))
for i in range(hparams.num_layers)
],
...
parser.add_argument(
'--first-layer-size',
help='Number of nodes in the 1st layer of the DNN',
default=100,
type=int
)
...
hptuning_config.yaml task.py
44. Proprietary + Confidential
In reality, ML is
Collect
data
Create the
model
Refine the
model
Understand
and prepare
the data
Serve the
model
Define
objectives
45. Proprietary + Confidential
Deploying the model
Creating model
gcloud ml-engine models create $MODEL_NAME --regions=$REGION
Creating versions
gcloud ml-engine versions create v1 --model $MODEL_NAME --origin $MODEL_BINARIES
--runtime-version 1.0
gcloud ml-engine models list
46. Proprietary + Confidential
Predicting
gcloud ml-engine predict --model $MODEL_NAME --version v1 --json-instances ../test.json
Using REST:
POST https://ml.googleapis.com/v1/{name=projects/**}:predict
JSON format (in this case):
{"age": 25, "workclass": "private", "education": "11th", "education_num": 7, "marital_status":
"Never-married", "occupation": "machine-op-inspector", "relationship": "own-child", "gender": "
male", "capital_gain": 0, "capital_loss": 0, "hours_per_week": 40, "native_country": "
United-States"}