Tech News

Unity Technologies has announced the release of their latest platform update Unity 2018.1 giving artists, developers and game engineers the power to express their talents and collaborate more efficiently to build games. Unity 2018.1 also marks the start of a new release cycle. Since 2017, Unity has adopted a new release plan where they come up with a new version every quarter and Unity 2018.1 marks the first version of the 2018 series. According to Brett Bibby, VP of Engineering, Unity Technologies, “With Unity 2018.1 we are introducing one of the largest upgrades in the history of our company, and it’s centered around two major concepts - next-level rendering and performance by default,” This release features two new upgrades: the Scriptable Render Pipelines and the Entity Component System. Together they make it easier for creators to make richer experiences utilizing modern hardware to deliver beautiful graphics. Next-level rendering with Scriptable Render Pipeline (SRP) Scriptable Render Pipeline (SRP) is available in the preview of Unity 2018.1. With SRP, developers and technical artists can now work directly with hardware and GPUs without having to go through millions of lines of C++ engine code. SRP makes it easy to customize the rendering pipeline via C# code and material shaders.   Unity 2018.1 also introduces two render pipelines. The High-Definition Render Pipeline (HD RP) is for developers with AAA aspirations. The Lightweight Render Pipeline (LW RP) is for those looking for a combination of graphics and speed. It optimizes the battery life for mobile devices and other similar platforms. Performance by default with the C# Job System &  Entity Component System (ECS) The C# Job system enables developers to write very fast, parallelized code in C# to take full advantage of multicore processors. It also provides protection from the pitfalls of multi-threading, such as race conditions and deadlocks. The runtime system is now combined with a new programming model, the Entity Component System. This new runtime system enables developers to use multicore processors without worrying about the programming. They can use this power to add more effects and complexity to games or add AI to make their creations richer and more immersive. It uses a data-oriented design instead of an object-oriented approach which makes it easier to reuse the code and easier for others to understand and work on it as well. Level design and shaders Unity 2018.1 reduces the time and effort required by artists, designers, and developers by allowing them to create levels, cinematic content, and gameplay sequences without coding. For this, new tools like ProBuilder/Polybrush and the new visual Shader Graph offer intuitive ways to design levels and create shaders without programming skills. ProBuilder is a unique hybrid of 3D-modeling and level-design tools optimized for building simple geometry, but capable of detailed editing and UV unwrapping as needed. With Polybrush developers can blend textures and colors, sculpt meshes and scatter objects directly in the Unity editor. Shader Graph can build shaders visually using a designer tool — without writing a single line of code. They offer easy drag-and-drop usability to create and connect nodes in a graph network. Unity Package Manager UI Unity 2018.1 builds on the package manager introduced in Unity 2017.2. It has a newly released Package Manager User Interface, the Hub, and Project Templates, to help start new projects faster and more efficiently. The Unity Package Manager UI  improves the following aspects of the project management workflow: Quick access to newly released features Get the latest fixes, instantly Access to Preview features Easily share lightweight projects Unity 2018.1 offers support for over 25+ platforms. This includes Magic Leap One, Oculus Go, ARCore 1.1, Android ARM64, Daydream Standalone and more. You can refer to the release notes for the full list of new features, improvements, and fixes. Unity will be showcasing all their latest innovations during Unite Berlin scheduled on June 19 - 21, 2018. Unity plugins for augmented reality application development Game Engine Wars: Unity vs Unreal Engine Unity releases ML-Agents v0.3: Imitation Learning, Memory-Enhanced Agents and more  

Microsoft had initially come up with Windows IoT which was formerly known as Windows Embedded. It was rebranded with the release of Windows 10 where Microsoft introduced twelve versions of Windows 10 that varied in features delivered, use cases, and the devices they supported. With that said, Microsoft gained a fighting place in the world of IoT with Windows 10 IoT which consists of two products catering to different customer bases: Windows 10 IoT Core and Windows 10 IoT Enterprise. Since IoT has to still evolve amongst major enterprises, we will focus on Window 10 IoT Core today. Windows 10 IoT Core is an optimized version of Windows 10 that is designed for smaller devices with or without a display that run on both ARM and x86/x64 devices. It is created to work on devices such as Raspberry Pi, Arduino, and other popular single board computers while it also utilizes the extensible Universal Windows Platform (UWP) API to build great solutions. The IoT domain has always been popular with traditional open source operating systems such as Linux distributions. Since the past couple of years, Windows has started to find its way into this domain and have proven to be an advantageous alternative in many ways. Initially setting up Windows 10 IoT Core to install the image and get started was a task. Recently Microsoft has focused on alleviating these small pain points and has got things sorted for Windows users. When it comes to developing IoT applications, open source distros lack making beautiful user interfaces possible. But with Windows this can be achieved thanks to Visual Studio. Visual Studio has always been a great environment to code in and if you are strong with C#, this can definitely be your go to platform. I emphasize on Windows users because  if you are looking at using or developing on Windows 10 IoT Core you would strictly need Windows 10 which isn’t open source. Well, this might never change. No doubt Microsoft wants to sell its software keeping its existing user happy. This would only be possible when Microsoft services work well only in its own environment. I’m sure you are wondering what could you possibly build with Windows 10 IoT Core and Raspberry Pi or Arduino. These are some breathtaking project ideas that you might be interested in building: Obstacle avoiding robot: This could be your basic project that can help you getting used to the new ecosystem you have adopted. Room light and temperature manager: Next, you can get some home automation tweaks that will help you automate your room environment.   Personal car data monitor: This can be an intermediate project where your IoT application can reveal the health of your vehicle before you start your ride.   Pet feeder: Lastly, you can take up something that involves Cloud platforms where you can feed your pet while your in office or at your neighbours instead of letting them starve. IoT is at such a stage now where the virtual world of Information Technology is connected to the read world. Initially this was possible only through Linux-based ecosystem, but with Windows 10 IoT coming into picture there has been quite a shift observed in the IoT market. Users have observed that in spite running on smaller devices Windows 10 IoT has managed to offer most of the essential features from parent Windows 10.  The world may still seem like a Linux base and deploying Python programs may look easier but it’s best to keep your options open and in this case you have a trusted platform, Windows. 5 reasons to choose AWS IoT Core for your next IoT project Should you go with Arduino Uno or Raspberry Pi 3 for your next IoT project? Splunk Industrial Asset Intelligence (Splunk IAI) targets Industrial IoT marketplace

Google Brain's Magenta project has released Magenta.js, a tool that could open up new opportunities in developing music and art with AI. The Magenta team have been exploring a range of ways to create with machine learning, but with Magenta.js, they have developed a tool that's going to open up the very domain they've been exploring to new people. Let's take a look at how the tool works, what the aims are, and how you can get involved. How does Magenta.js work? Magenta.js is a JavaScript suite that runs on TensorFlow.js, which means it can run machine learning models in the browser. The team explains that JavaScript has been a crucial part of their project, as they have been eager to make sure they bridge the gap between the complex research they are doing and their end users. They want their research to result in tools that can actually be used. As they've said before: "...we often face conflicting desires: as researchers we want to push forward the boundaries of what is possible with machine learning, but as tool-makers, we want our models to be understandable and controllable by artists and musicians." As they note, JavaScript has informed a number of projects that have preceded Magenta.js, such as Latent Loops, Beat Blender and Melody Mixer. These tools were all built using MusicVAE, a machine learning model that forms an important part of the Magenta.js suite. The first package you'll want to pay attention to in Magenta.js is @magenta/music. This package features a number of Magenta's machine learning models for music including MusicVAE and DrumsRNN. Thanks to Magenta.js you'll be able to quickly get started. You can use a number of the project's pre-trained models which you can find on GitHub here. What next for Magenta.js? The Magenta team are keen for people to start using the tools they develop. They want a community of engineers, artists and creatives to help them drive the project forward. They're encouraging anyone who develops using Magenta.js to contribute to the GitHub repo. Clearly, this is a project where openness is going to be a huge bonus. We're excited to not only see what the Magenta team come up with next, but also the range of projects that are built using it. Perhaps we'll begin to see a whole new creative movement emerge? Read more on the project site here.

Angular 6 finally arrives! This is a major production release of Angular, the popular JavaScript framework for building web and mobile applications. This release mainly focuses on the toolchain and on making it easier for developers to migrate to future versions of Angular quickly. With this release, major framework packages (@angular/core, @angular/common, @angular/compiler, etc), the Angular CLI, and Angular Material + CDK are also synchronizing their releases. All are releasing as 6.0.0 today. Here’s a quick rundown of all major features: New CLI commands Two new CLI commands have been added. The ng-update command recommends updates to an application by analyzing the package.json. ng-update will help developers adopt the right version of dependencies while keeping them in sync. The ng-add CLI command adds new capabilities to a project by using the package manager to download new dependencies and invoke an installation script. ng add @angular/pwa —Converts your app into a PWA by adding an app manifest and service worker ng add @ng-bootstrap/schematics — Adds ng-bootstrap to your application ng add @angular/material — Install and setup Angular Material and theming and register new starter components into ng generate CLI Workspaces CLI v6, which is a part of Angular 6 release, now supports workspaces containing multiple projects, such as multiple applications or libraries. CLI projects will now use angular.json instead of .angular-cli.json for build and project configuration. It also adds support for creating and building libraries with the command ng generate library <name>. This command will create a library project within the CLI workspace, and configure it for testing and building. Angular Elements Angular 6 also comes with the first release of Angular Elements. Angular elements allow bootstrapping Angular components within an existing Angular application by registering them as Custom Elements. They replace the need to manually bootstrap Angular components found in static html content. Angular Material + CDK Components Angular 6 features a new tree component for displaying hierarchical data. The Tree component in Angular Material and the Component Dev Kit helps in better visualization of tree structures such as a list of files. Alongside the tree, there are new badge and bottom-sheet components. Badges help display small bits of helpful information, such as unread item counts. Bottom-sheets are a special type of mobile-centric dialogs, commonly used to present a list of options following an action. With the release of v6, the @angular/cdk/overlay package includes new positioning logic that helps make pop-ups which remain on-screen in all situations. The angular material also includes 3 new starter components. Material Sidenav: Generates a starter component including a toolbar with the app name and the side navigation. Material Dashboard: Generates a starter dashboard component containing a dynamic grid list of cards. Material Data Table: Generates a starter data table component that is pre-configured with a datasource for sorting and pagination. Updated to use RxJS v6 Angular has been updated to use RxJS v6. RxJS v6 was introduced at ng-conf and brings several major changes, along with a backwards compatibility package rxjs-compat for keeping applications working without breaking components. Long Term Support Expansion The angular community has extended the long-term support to all major releases starting with v4. Each major release will be supported for 18 months with around 6 months of active development followed by 12 months of critical bug fixes and security patches. A common complaint among developers about Angular has been about the messy migrations from one version to another. This announcement aims to make updating from one major to the next easier, and give bigger projects more time to plan updates. How can you upgrade to the new version? The update will take advantage of the new ng update tool. Here are the steps for updating. Update @angular/cli Update your Angular framework packages Update other dependencies Checkout the Angular blog for detailed release notes and steps on how to update. ng-conf 2018 highlights, the popular angular conference Why switch to Angular for web development – Interview with Minko Gechev 8 built-in Angular Pipes in Angular 4 that you should know  

PyTorch has announced a preview of the blueprint for PyTorch 1.0, the next major release of the framework. This breakthrough version is expected to bring more stability, integration support and complete production backing allowing developers to move from core research to production in an amicable way without having to deal with any migration challenges. PyTorch is an open-source Python-based scientific computing package which provides powerful GPU acceleration. PyTorch is known for advanced indexing and functions, imperative style, integration support and API simplicity. This is one of the key reasons why developers prefer PyTorch for research and hackability. To know more about how Facebook-based PyTorch competes with Google’s TensorFlow read our take on this deep learning war. Some of the noteworthy changes in the roadmap for PyTorch 1.0 are: Production support One of the biggest challenges faced by developers in terms of using PyTorch is production support. There are n number of issues faced while trying to run the models efficiently in production environments. Even though PyTorch provides excellent simplicity and flexibility, due to its tight coupling to Python, the performance at production-scale is a challenge.   To counter these challenges, the PyTorch team has decided to bring PyTorch and Caffe2 together to provide production-scale readiness to the developers. However, adding production support brings complexity and configurable options for models in the API. The PyTorch team will stick to the goal of keeping the platform -- a favorable choice -- for researchers and developers. Hence, they are introducing a new just-in-time (JIT) compiler, named torch.jit. torch.jit compiler rewrites PyTorch models during runtime in order to achieve scalability and efficiency in production environments. It can also export PyTorch models to run in a C++ environment. (runtime based on Caffe2 bits) Note: In PyTorch version 1.0, your existing code will continue to work as-is. Let’s go through how JIT compiler can be used to export models to a Python-less environment in order to improve their working performance. torch.jit: The go-to compiler for your PyTorch models Building models using Python code, no doubt gives maximum productivity and makes PyTorch very simple and easy-to-use. However, this also means PyTorch finding it difficult to know which operation you will run next. This can be frustrating for the developers during model export and automatic performance optimizations because they need to be aware of how the computations will look like before it even gets implemented. To deal with these issues, PyTorch provides two ways of recovering information from the Python code. Both these methods will be useful based on different contexts, giving you the leverage to use/mix them with ease. Tracing the native Python code Compiling a subset of the Python language Tracing mode torch.jit.trace function allows you to record the native PyTorch operations performed along with the data dependencies between them. PyTorch version 0.3 already had a tracer function which is used to export models through ONNX. This new version uses a high-performance C++ runtime that allows PyTorch to re-execute programs for you. The key advantage of using this method is that it doesn’t have to deal with how your Python code is structured since we only trace through native PyTorch operations. Script mode PyTorch team has come up with a solution called scripting mode made specially for those models such as RNNs which make use of control flow. However, you will have to write out a regular Python function (avoiding complex language features) In order to get your function compiled, you can assign @script decorator. This will make sure it alters your Python function directly into high-performance C++ during runtime. Advantages in optimization and export techniques Irrespective of you using a trace or a script function, the technique allows you to optimize/export the model for use in production environments (i.e. Python-free portrayal of the model) Now you can derive bigger segments of the model into an intermediate representation to work with sophisticated models. You can use high-performance backends available in Caffe2 to run the models efficiently Usability If you don’t need to export or optimize your model, you do not need to use these set of new features. These modes will be included into the core of the PyTorch ecosystem, thus allowing you to mix and match them with the existing code seamlessly as per your needs. Additional changes and improvements In addition to the major update in the production support for 1.0, PyTorch team will continue working on optimizing, working on the stability of the interface, and fixing other modules in PyTorch ecosystem PyTorch 1.0 will see some changes in the backend side which might affect user-written C and C++ extensions. In order to incorporate new features and optimization techniques from Caffe2, PyTorch team is replacing (optimizing) the backend ATen library. PyTorch team is planning to release 1.0 during the summer. For the detailed preview of the roadmap, you can refer the official PyTorch blog. Top 10 deep learning frameworks The Deep Learning Framework Showdown: TensorFlow vs CNTK Why you should use Keras for deep learning

KubeCon and Cloud Native Computing Foundation (CNCF) have been running a joint summit in Copenhagen this week. There has been a whole host of updates and announcements from some of the biggest cloud vendors, from Oracle to Google. That's important as it highlights that Kubernetes has well and truly established itself within the container space. Months after Docker conceded ground to the project in the orchestration world, vendors are looking to adapt to Kubernetes status on today's software landscape. 5 important vendor announcements from KubeCon + CloudNativeCon Let's take a look at some of the biggest announcements from KubeCon and CNC and what they mean for the industry. Oracle Oracle has made a number of announcements in Copenhagen that underline not only the dominance of Kubernetes, but the growth of serverless computing as well. The organization's Fn Project, Oracle's serverless cloud project, are working closely with Cloud Native Computing Foundation to develop open standards. This includes support for the Cloud Events initiative, which aims to standardize how event data is described. Oracle also revealed it was launching a container engine for Kubernetes. Oracle Container Engine has been developed to help Oracle's customers tackle a range of common infrastructure challenges, such as security and networking. Both announcements highlight the changing needs of Oracle's customers. It also underscores how open source software is transforming the way established vendors act and view the world. They need to adapt. Google Google announced gVisor. gVisor is a runtime environment that allows you to separate containerized applications from the kernel on which they are based. The company also revealed Stackdriver Kubernetes Monitoring. This is an interesting tool as it should simplify the way in which you monitor Kubernetes on the Google cloud platform. Essentially, it brings various different components into one place. You'll now be able to see a range of metrics and events across containers and clusters. Cloud66 Cloud 66 introduced a number of new features designed to enhance Skycap, its flagship container delivery pipeline product. Stencils is, as the name suggests a way of templating Kubernetes configuration files. This will make managing accessibility to those files easier, and means that making changes won't impact releases in the way they might otherwise. Formations, meanwhile, allow you to target container deployments to particular clusters. Cloud 66 also revealed an open source tool called Copper. Copper validates Kubernetes configuration files; it's essentially a way of allowing you to test and check the permissions and overall configuration of the files. In the press release, CEO Khash Sajadi said: "With the advance of micro-services, containers and the surge of APIs, developers and operations teams appreciate a self-service toolchain that operations curate, and developers can run with in production. Cloud 66 is committed to tools that provide a balance between operational governance and development freedom, in the cloud or for on-premises deployments." Cisco Cisco used KubeCon to reveal a couple of important Kubernetes-related updates to two of their products. AppDynamics, the application performance analytics tool, and CloudCenter, both now have Kubernetes support. This move will bring Kubernetes into many legacy applications that have previously been locked into the level of functionality offered by Cisco. Here's what Kip Compton, the VP of Cisco Cloud Platform and Solutions Group had to say: "The Kubernetes platform has emerged as the de-facto container solution as customers accelerate adoption of containerized application architectures... But organizations are still challenged to efficiently and confidently utilize Kubernetes as they modernize legacy applications and develop new cloud applications. With our latest Kubernetes support, customers can now easily adopt production-grade Kubernetes across multicloud environments.” This is interesting - Compton identifies a common challenge around bringing legacy software up to date. With this announcements, Cisco is helping their customers to find a way around legacy issues, reducing the need to undergo a risky mass system migration. Digital Ocean Cloud platform Digital Ocean released a Kubernetes product in Copenhagen. Like the Cisco release, at a most basic level, it's going to make it much easier for engineering and operations teams to leverage Kubernetes without the challenges faced with integrating the various platforms. Learn more about Digital Ocean Kubernetes here. Read next Google’s kaniko – An open-source build tool for Docker Images in Kubernetes, without a root access Kubernetes 1.10 released The key differences between Kubernetes and Docker Swarm

Graph database project Neo4j has announced the release of Neo4j 3.4. The update is poised to strengthen Neo4j's position as the market leader in the world of connected data, and could make it more accessible for more organizations and users. Alongside the updates and improvements in Neo4j 3.4, the team have also announced Neo4j Bloom. This is a data visualization tool designed to make it easier to communicate and present insights to key stakeholders. It's a well-established fact that communication is key if you're going to do data science and data analysis well, so it's clear that Neo4j are directly responding to a key issue for their customers. It's also a valuable hook for any potential new customers. Emil Eifrem (@emileifrem) founder and CEO of Neo4j has this to say about the updates: "Our investments in the Neo4j database extend its ability to scale and drive new use cases for our current customers... Graph databases are an enterprise standard, and the introduction of Neo4j 3.4 and Neo4j Bloom means more people will discover the power and value of connected data." What's new in Neo4j 3.4? There are a number of important new features in Neo4j 3.4. Let's take a look at some of them. There have been some impressive improvements in performance. Neo4j has always been best in class when it comes to the performance of its database. But with 3.4, it consolidates its position. Cypher now executes 70% faster, data loading is now 30%-50% faster, and backups are apparently now 100% faster. Multi-clustering brings fully sharded horizontal scaling one step closer. 3D geospatial and data/time search functions extend the range of potential use cases. That means the graph databases users build have additional dimensions for users to search. Neo4j Bloom: making connected data accessible Neo4j is a tool that has been designed by the project to make complex graph databases easier to visualize and explore. Abstract relationships that may be difficult to interpret and understand for an outsider will become much clearer. That's good news for data scientists and data analysts, but it's also good news for Neo4j - key stakeholders and decision makers who don't have technical expertise will now also be paying attention to Neo4j. It's important to note the uniqueness of Neo4j Bloom. It is able to present the connections and context around various data points; that gives it an extra dimension over other popular data visualization tools. "Neo4j Bloom is specifically designed to illuminate connections between data points in an intuitive way, especially for executives and stakeholders who might not be very technical" explains Eifrem. You can find out more about Neo4j 3.4 here. From Graph Database to Graph Company: Neo4j’s Native Graph Platform addresses evolving needs of customers Creating a graph application with Python, Neo4j, Gephi & Linkurious.js

The 2018 angular conference (ng-conf 2018) took place on April 18–20th 2018 at Salt Lake City, UT. The conference featured a large number of sessions, workshops, and speakers from the Angular team and the Angular community. ng-conf 2018 was live streamed and live transcripted for the home audience, to enjoy the same learning experiences as those of the actual attendees. Not to mention, the whole event was 80s themed, coinciding with the release of the movie Ready Player One which features a lot of 1980s pop-culture references. We have compiled a list of popular announcements and sessions which were the highlights of this year’s conference. Introducing RxJS6 Ben Lesh introduced version 6 of the ReactiveX library for JavaScript. RxJS is a library for reactive programming using Observables, that makes it easier to compose asynchronous or callback-based code. RxJS6 brings cleaner imports while having a smaller API, a backward compatibility package to update without changing your code, and automatic code migration for TypeScript. RxJS 6 Backward Compatibility: To make the migration path from RxJS 5 to RxJS 6, the RxJS team has released a sibling package called rxjs-compat. This package creates a compatibility layer between the APIs of v6 and v5. Automatic code migration: For TypeScript users, which cover the majority of Angular developers, RxJS introduces tslint, which offers a great deal of automated refactoring to make the transition from v5 to v6 even easier. Deprecations: A large no. of deprecations have been made. This includes Result selectors, Observable.if (replaced by iif() and Observable.throw replaced by throwError().  Apart from this, other deprecated methods include merge, concat, combineLatest, race, and zip. StackBlitz + Angular: A Better Way to Build PWA’s Albert Pai and Eric Simons conducted a session on building PWAs using StackBlitz with Angular. This suite of new developer tools makes building and debugging progressive web apps a lot easier. They run entirely in your browser with no setup or configuration required. NgRx Sessions A lot of sessions revolved around NgRx, the RxJS powered state management for Angular applications, inspired by Redux. Brandon Roberts talked about how to implement authentication with a reactive store by building an auth-based app with NgRx Store, Router, and Effects. Brandon Roberts and Mike Ryan presented a talk on Reactive Testing Strategies with NgRx. He talked about testing strategies such as unit testing presentation components, integration testing with smart components, testing observables, state management, and end-to-end tests, to make testing a reactive application easier and to simplify the testing triangle. Vitalii Bobrov talked about how NgRx Schematics is a huge time-saver. It will automate NgRx code generation and give you the ability to focus on application business logic. Mike Ryan presented a talk on Good Action Hygiene with NgRx and talked about how to write clean actions and avoid common anti-patterns. Jesse Sanders talked about how to handle complex forms using ngrx. David East and Todd Motto conducted a workshop on NgRx Selectors. They talked about multiple benefits of Selectors are easy to create and work well with teams. They allow you compute state from your store, which acts like a view model. Selectors are easily tested, memoized for performance, and compose-able for re-usability. By the end of this talk they promised to shake any store structure fears developers may have so that they can move forward boldly with selectors. Firebase, Cloud Functions, and Machine Learning Jason Dobry presented a workshop on how to use the Firebase SDK for Google Cloud Functions to improve an AngularFire Chat Web app. AngularFire is the officially supported AngularJS binding for Firebase. This binding lets you associate Firebase references with Angular models so that they will be transparently and immediately kept in sync with the database and with all other clients currently using your application. Jason also talked about how to use Cloud Functions to send notifications to users of the Chat app, use the Google Cloud Vision API to process images, and use the Google Natural Language API to process chat messages. The conference also featured workshops on the following How to hack an Angular app, Writing A Custom Angular Build Make your JS app search engine friendly, Building components with the Angular Component Dev Kit Strategies for Server Side Rendering Angular Apps Angular Elements in v6 and Beyond Hands-on Full-Stack development with Nx and Bazel, Using StackBlitz & Angular for Rapid App Prototyping Reusable Animations, Google’s serverless tools, VR Hero, and more.   You can have a look at the entire list of sessions and workshops on the ng-conf website. Why switch to Angular for web development – Interview with Minko Gechev 8 built-in Angular Pipes in Angular 4 that you should know Building Components Using Angular

At the two-day F8 conference hosted by Facebook, Oculus unveiled a new virtual reality headset. The Oculus Go is priced at an astonishing $199, a lot lesser than its predecessors. (Oculus Rift headset costs around $399). Here’s a quick rundown of all the key features. Self-contained headset The Oculus Go is completely self- contained and stand alone. Everything including the hardware, screen, and processor, is contained within the headset. The functional pistol-grip Oculus controller is also included in the box. Developers don’t need a special computer, graphics card, game console or even a phone to operate the VR device as it is completely autonomous. Rich Display Oculus is equipped with a 5.5-inch, 2,560x1,440-pixel LCD display that looks particularly crisp when reading text or watching videos.  It uses optimized 3D graphics which reduced the screen-door effect typically encountered on most VR headsets. It uses fixed foveated rendering, rendering the area at the center of the display more sharply than the edges, to make many apps look even better. Powerful Sound Spatial audio drivers are built into the headset which provides direct, immersive, surround sound, without the need for earphones. Alternatively, it also has a 3.5mm audio jack. Lightweight and Comfortable The Oculus Go is comfortable and well designed. The Go goggles have breathable fabrics, injection foam molding, and other advances in wearable materials for better comfort. It is also lightweight and portable. For all the great features, the product is not without its limitations. The screen has a narrower field of view (FOV) than Oculus Rift and HTC Vive. It also does not include a slider or scalar to adjust "interpupillary distance", i.e. how images line up with your own face. Oculus Go only recognizes three degrees of freedom (3DOF). So, realistic VR effects remain when you rotate or tilt your head. However, the effect is broken as you lean in any direction. Go does not offer positional tracking while seated or while walking.   Nevertheless, Oculus currently supports over 1,000 existing apps, and pairs with both iPhones and Android phones, making it one of the best iPhone VR headsets around right now.  The pricing is set to be $199 USD for the 32GB model and $249 for the 64GB version. Consumers can now purchase the headset via the Oculus website in 23 countries. Facebook’s F8 Conference – 5 key announcements Understanding the hype behind Magic Leap’s New Augmented Reality Headsets Leap Motion open sources its $100 augmented reality headset, North Star

Nvidia Tesla V100 GPUs are now publicly available in beta on Google Compute Engine and Kubernetes Engine. Also, Nvidia Tesla P100 GPUs are now generally available. Nvidia Tesla V100 GPU is almost equal to 100 CPUs. This gives customers more power to handle computationally demanding applications, like machine learning, analytics, and video processing. One can select as many as eight NVIDIA Tesla V100 GPUs, 96 vCPU and 624GB of system memory in a single VM, receiving up to 1 petaflop of mixed precision hardware acceleration performance. NVIDIA V100s are available immediately in the following regions: us-west1, us-central1 and europe-west4. Each V100 GPU is priced as low as $2.48 per hour for on-demand VMs and $1.24 per hour for Preemptible VMs. Making Nvidia Tesla V100 available on the compute engine is part of Google’s GPU expansion strategy. Similar to Google GPUs, the V100 is also billed by the second and Sustained Use Discounts apply. NVIDIA Tesla P100 GPU, on the other hand is a good fit if one wants a balance between price and performance. One can select up to four P100 GPUs, 96 vCPUs and 624GB of memory per virtual machine. The P100 is also now available in europe-west4 (Netherlands) in addition to us-west1, us-central1, us-east1, europe-west1 and asia-east1. * Maximum vCPU count and system memory limit on the instance might be smaller depending on the zone or the number of GPUs selected. ** GPU prices listed as hourly rate, per GPU attached to a VM that are billed by the second. Pricing for attaching GPUs to preemptible VMs is different from pricing for attaching GPUs to non-preemptible VMs. Prices listed are for US regions. Prices for other regions may be different. Additional Sustained Use Discounts of up to 30% apply to GPU on-demand usage only. Google Cloud makes managing GPU workloads easy for both VMs and containers by providing, Google Compute Engine where customers can use instance templates and managed instance groups to easily create and scale GPU infrastructure. NVIDIA V100s and other GPU offerings in Kubernetes Engine, where Cluster Autoscaler helps provide flexibility by automatically creating nodes with GPUs, and scaling them down to zero when they are no longer in use. Preemptible GPUs for both Compute Engine managed instance groups and Kubernetes Engine’s Autoscaler optimizes the costs while simplifying infrastructure operations. Read more about both the GPUs in detail on the Google Research Blog and benefits of each on Nvidia V100 and Nvidia P100 blog post. Google announce the largest overhaul of their Cloud Speech-to-Text Google’s kaniko – An open-source build tool for Docker Images in Kubernetes, without a root access How machine learning as a service is transforming cloud  

The ICLR paper, ‘Zero-Shot learning for Visual Imitation’ is a collaborative effort by Deepak Pathak, Parsa Mahmoudieh, Michael Luo, Pulkit Agrawal, Dian Chen, Fred Shentu, Evan Shelhamer, Jitendra Malik, Alexei A. Efros, and Trevor Darrell. In this article, we will come across one of the main problems with imitation learning, the expense of expert demonstration. The authors here propose a method for sidestepping this issue by using the random exploration of an agent to learn generalizable skills which can then be applied without any specific pretraining on any new task. Reducing the expert demonstration expense with Zero-shot visual imitation What problem is the paper trying to solve? In order to carry out imitation, the expert should be able to simply demonstrate tasks capably without lots of effort, instrumentation, or engineering. Collecting too many demonstrations is time-consuming, exact state-action knowledge is impractical, and reward design is involved and takes more than task expertise. The agent should be able to achieve goals based on the demonstrations without having to devote time learning to do each and every task. To address these issues, the authors recast learning from demonstration into doing from demonstration by (1) Only giving demonstrations during inference and, (2) Restricting demonstrations to visual observations alone rather than full state-actions. Instead of imitation learning, the agent must learn to imitate. This is the goal that the authors are trying to achieve. Paper summary This paper explains how existing approaches to imitation learning distill both what to do (goal) and how to do it (skills), from expert demonstrations. However, this expertise is effective but expensive supervision: it is not always practical to collect many detailed demonstrations. The authors here suggest that if an agent has access to its environment along with the expert, it can learn skills from its own experience and rely on expertise for the goals alone. And so, they have proposed a ‘Zero-shot’ method which does not include any expert actions or demonstrations during learning. The zero-shot imitator has no prior knowledge of the environment and makes no use of the expert during training. It learns from experience to follow experts, for instance, the authors conducted certain experiments such as, navigating an office with a turtlebot, and manipulating rope with a baxter robot. Key takeaways The authors have proposed a method for learning a parametric skill function (PSF) that takes as input a description of the initial state, goal state, parameters of the skill and outputs a sequence of actions (could be of varying length), which take the agent from initial state to goal state. The authors have shown real-world results for office navigation and rope manipulation but make no domain assumptions limiting the method to these problems. Zero-shot imitators learn to follow demonstrations without any expert supervision during learning. This approach learns task priors of representation, goals, and skills from the environment in order to imitate the goals given by the expert during inference. Reviewer comments summary Overall Score: 25/30 Average Score: 8 As per one of the reviewers, the proposed approach is well founded and the experimental evaluations are promising. The paper is well written and easy to follow. The skill function uses a RNN as function approximator and minimizes the sum of two losses i.e. the state mismatch loss over the trajectory (using an explicitly learnt forward model) and the action mismatch loss (using a model-free action prediction module) . This is hard to do in practice due to jointly learning both the forward model as well as the state mismatches. So first they are separately learnt and then fine-tuned together. One Shot Learning: Solution to your low data problem Using Meta-Learning in Nonstationary and Competitive Environments with Pieter Abbeel What is Meta Learning?

This year's F8 Conference is one of the most closely watched ever, but not necessarily for the right reasons. Thanks to the Cambridge Analytica scandal and Mark Zuckerberg getting flustered in front of Congress, the spotlight is shining brightly over Facebook and its CEO. Zuckerberg's keynote speech was a focal point. Would he apologize? What would the big announcements be? In truth, the keynote was actually pretty interesting. There were some big surprises and interesting developments. Notably absent, however, was any form of apology from Zuckerberg. The MIT Technology Review were particularly incisive on this: '“Safety,” yes; “sorry,” no' runs the headline on their F8 Conference analysis. So while the F8 keynote speech may have lacked an apology, it was an interesting exercise in avoiding the issue. There was plenty of positivity, and lots of new initiatives that indicate that Facebook remains an optimistic and forward thinking organization. Either that, or labouring under considerable seld-deception - that depends on your viewpoint. 5 announcements from Facebook's F8 Conference you need to know So, let's take a look at 5 of the biggest announcements from Facebook's F8 Conference and what they might mean. Facebook is going to launch a dating app What's the best way to say sorry? Build a dating app! That way you can detract attention from the problems your product may or may not be causing civic society as people begin to get distracted by ill-advised crushes and unnecessary flirting. There are a number of ways you could view this announcement. On the one hand it looks a lot like Hubris, trying to beat established dating apps like Tinder at their own game. From this angle, it certainly looks like a feint to distract people from some of the thornier issues around the platform. But on the other hand it might just be a genius move from Zuckerberg. Literally billions of people live their lives on Facebook; why shouldn't their love lives be a part of that too? When you consider that the share value of Match Group - the organization that owns Tinder and OKCupid - took a tumble, Facebook might just be on to a winner to get itself out of a hole. Facebook is going to let you clear your history Although much of the commentary on Zuckerberg's F8 Conference keynote speech has been critical of the Facebook chief, one announcement did indicate that perhaps Facebook is actually learning its lessons. "Clear History" is a feature that does exactly what you think it does. It allows you to wipe any data that has been collected by sites and third party applications. Zuckerberg was keen to warn that this feature might make the experience 'worse' for users. This is what he said: "When you clear your cookies in a browser, you may have to sign back into dozens of websites... The same is going to be true here. Your Facebook won’t be quite as good as it relearns your preferences. But after going through our systems, this is the kind of control we think people should have.” It will be interesting to see whether users actually do want this level of control, or whether they're happy to give up some degree of privacy for greater personalization and convenience. The Facebook app review process is coming back By contrast, this announcement from Zuckerberg looks a lot more like an attempt to draw a line under the problems have been facing over the last few months. After closing the app review process following the Cambridge Analytica scandal, it was revealed that third party apps and chatbots would be able to go through the process once again. It's not clear how much this process will have changed and whether any other forms of governance will be in place - that remains to be seen. However, Zuckerberg didn't reveal a date as to when this might happen. Whether that means they need more time to refine the process or they're simply delaying it, it suggests that this announcement was as much to strike a tone of positivity with the developers in attendance. Facebook is simplifying the messenger app Anyone who has used Facebook messenger will know how irritating it can be too use. It certainly isn't minimal, and it would be hard to describe it as slick. This is something Zuckerberg seemed to appreciate, saying "when you’re messaging, you really want a simple and fast experience." This announcement was a nice counterpoint to news around the dating app. Just as it looks like Facebook has finally thrown the kitchen sink and everything else into the user experience, Zuckerberg takes a step back with a clear acknowledgement of user needs. A clean and slick application might not make you forgive him for undermining the foundations of western democracy, but it does make the bitter pill of reality in 2018 a little easier to swallow. Oculus Go is now available for $199 The VR headset market has been quietly growing. With the Oculus Go, Facebook have a product that might just capture the attention of millions of people who haven't yet felt comfortable with the options out there. Not only is it relatively cheap at $199, it's been described as a "game-changer" by TechCrunch. The reviewer writes that the headset "minimizes friction and promotes a surprisingly capable baseline for bringing new consumers into VR for the first time." With VR memories also touted to soon hit news feeds, Facebook might just be on course to make VR mainstream. Despite the hype, no one has been able to do that yet. F8 Conference: was Zuckerberg papering over the cracks? Clearly, there was plenty to get excited about in Zuckerberg's keynote speech. Whether you're a Facebook user or developer, there's enough to make you think there's still plenty of life and confidence in the platform yet. Perhaps rumours of its slow death have been exaggerated. But you can't deny there are problems at Facebook. A dating app isn't going to fix them. Oculus Go isn't going to transform the fortunes of the company overnight. What Facebook really needs is a frank conversation about what it does well. And that, really, is about delivering a platform users can't help but use. Once this year's F8 Conference is over, maybe that conversation can happen. Until then, let's just watch what happens over the next few days.

After the recent release of Node 10.0.0, npm have released version 6 in collaboration with node.js. npm v6 is a major update of the popular package manager for the JavaScript runtime environment Node.js. Typically, npm release their newer versions every year around spring time and following this pattern npm v6 was introduced as on April 26, 2018. This update introduces powerful security features for every developer who works with open source code. Built in security features npm v6 is the result of the collaboration between npm and their acquisition of the Node Security Platform. This introduces two new security features: npm registry Every user of the npm v6 Registry will begin receiving automatic warnings if the code used has a known security issue. npm will automatically review install requests against the NSP database and return a warning if the code contains a vulnerability. npm audit npm v6, has a new command, ‘npm audit’, which allows developers to recursively analyze their dependency trees to identify specific insecurities, following which developers can swap in a new version or find a safer alternate dependency. Both these security features are available free of charge to every npm user, with no purchase or registration required. These resources are open sourced to maximize the community benefit. By alerting the entire community to security vulnerabilities within a tool, npm can make JavaScript development safer for everyone. Additional Features Apart from the security features, there are also a large number of other performance updates: npm v6 is up to 17x faster than the npm of one year ago. npm ci is optimized to use npm within the continuous integration/continuous deployment (CI/CD) workflow almost 2x–3x faster. Webhooks are now configurable directly within the npm CLI. Easy verification of package with respect to tampering and corruption, with more visibly integrated metadata. Teams can now more easily share reproducible builds with automatic resolution of lockfile conflicts. Also checkout the release notes for npm v6 release, and the roadmap of the year ahead. Node 10.0.0 released, packed with exciting new features How is Node.js Changing Web Development? How to deploy a Node.js application to the web using Heroku

Infrastructure made of connected things is highly trending as organizations are in the process of adopting Internet of Things. At the same time security concerns around these connected devices continues to be a bottleneck for IoT adoption. In an effort to improve IoT security, earlier this month, Microsoft released Azure Sphere, a cost-effective way of securing connected devices. Gartner claims that worldwide spending on IoT security will reach 1.5 billion in 2018. Azure Sphere is basically a suite of services, used to enhance IoT security. Following are the services included in the suite: Azure Sphere MCUs These are a certified class of microcontrollers specially designed for security of internet of things. It follows a cross-over mechanism which allows the combination of running realt-time and application processors with built-in microsoft security mechanism and connectivity. MCU chips are designed using custom silicon security technology, made by Microsoft. Some of the highlights are: A pluton security subsystem to execute complex cryptographic operations A cross-over MCU with the combination of both Cortex-A and Cortext M class processor. Build-in network connectivity to ensure devices are upto date Azure Sphere OS Azure Sphere OS is nothing but a Linux distro used to securely run the internet of things. This highly scalable and secure operating system can be used to run the specialized MCUs by adding an extra layer of security. Some of the highlights are: Secured application containers focussing on agility and robustness A custom Linux Kernel enabling silicon diversity and innovation A security monitor to manage access and integrity The Azure Sphere Security Service An end-to-end security service solely dedicated to secure Azure sphere devices, enhancing security, identifying threats, and managing trust between cloud and device endpoints. Following are the highlights: Protects your devices using certificate based-authentication system. Ensure devices authenticity by ensuring that they are running on genuine software Managing automated updates for Azure Sphere OS, for threat and incident response Easy deployment of software updates to Azure Sphere connected devices. For more information, refer the official Microsoft blog. Serverless computing wars: AWS Lambdas vs Azure Functions How to call an Azure function from an ASP.NET Core MVC application

Slack: productivity and collaboration tool, or platform for procrastination, in-jokes and GIFs? We couldn't possibly say here at Packt. For most of us, the only thing worse than wasting time on Slack is looking like you're never on Slack at all. While you'd like to tell people it's because you're busy, you can see your colleagues eyeing you with suspicion, convinced that if you're not procrastinating in the same manner they are, you really can't be doing anything at all. Luckily, someone has invented a tool for dealing with exactly this problem.  Take a bow Will Leinweber (@leinweber) - you have made something to make us look busy. Or, at the very least thoughtful and ready to contribute to the channel chat at any moment. https://twitter.com/leinweber/status/989267343002951680 Will has put the project on GitHub. You can find it here. The only disappointment with the tool is that Will didn't include the additional feature that  "asks other people what their typing whenever they're typing." The results were pretty hilarious, and likely too distracting for anyone to do any work at all... https://twitter.com/leinweber/status/989285775165423616 Needless to say, there was a pretty strong reaction to Will's program. https://twitter.com/snail_5/status/989271471766757376 https://twitter.com/LittleMxSurly/status/989315676325085184 https://twitter.com/CodeTheWebBlog/status/990008655394189313 https://twitter.com/shandrew/status/989395097249693698 Truly, software is being used for incredible things in 2018. These are the projects we need if we're to survive a hostile and unforgiving future, forever typing into the abyss at each other, and doomed to search out reaction GIFs to every rude email and hostile expression someone sends your way. What other novelty software projects have you seen recently? Let us know in the comments, and we'll do some investigative work* *Have a look on Twitter. Read more Creating slash commands for Slack using Bottle