Attend FREE Webinar on Data Science for Career Growth Register Now

Google’s Cloud AutoML Vision: Train Machine Learning Models Without Coding

5 (100%) 6 votes

Even with consistent data and Machine Learning Engineers on the go, using primary techniques to build the reliable and appropriate training model for your dataset is very laborious. With a comparatively few number of skilled individuals that can fill those jobs, demand for Artificial Intelligence or Machine Learning experts is more than ever right now. And most importantly, those folks don’t come cheap.

Announced at company’s I/O conference last year, Google on 17th declared the Alpha launch of Cloud AutoML. A new facility that enables developers- including those with no ML expertise- build custom Image Recognition models.

Data Analytics Course by Digital Vidya

Free Data Analytics Webinar

Date: 27th Jun, 2019 (Thursday)
Time: 3 PM (IST/GMT +5:30)

Currently limited to Image recognition, we can expect the company to, later on, launch the similar versions for other Machine Learning building blocks in its stack (i.e. Speech Recognition, Video Analysis, Text analysis, Translator etc.).

According to Google, the basic idea about Cloud AutoML here is to allow anybody to bring their images, upload them and then have Google’s systems automatically create a high-quality customer machine learning model for them. Which is being done with help of Google’s art transfer learning and Neural technology search technology.

In an official blog post, Google’s Chief Scientist of Cloud AI, Fei-Fei Li said: “Our first Cloud AutoML release will be Cloud AutoML Vision, a service that makes it faster and easier to create custom ML models for image recognition. Its drag-and-drop interface lets you easily upload images, train and manage models, and then deploy those trained models directly on Google Cloud. Early results using Cloud AutoML Vision to classify popular public data sets like ImageNet and CIFAR have shown more accurate results with fewer misclassifications than generic ML APIs.”

Disney is one of the early customers of Cloud AutoML technology. Mike White, CTO, SVP, Disney Consumer Products and Interactive Media said that this technology is helping them to annotate their products with Disney products and improvise guest experience through relevant search results. Shopping and user experience is one way to look at it! Another early customer is ZSL. ZSL is an international conservation charity mainly concerned about conservation of animals and their habitats. In order to do that firstly, they have to track and analyze the current scenario. They have placed series of cameras that take pictures of passing animals when triggered by heat or motion. And here to analyze those images, they have been using Google’s Cloud AutoML technology.

How Cloud AutoML works

Peer comparison

Google points out that AutoML vision is one of its kind but there are similar services that take the similar approach (i.e. Around three months back, the pioneer in cloud market Amazon Web Services introduced SageMaker for making and launching AI models. Same as that, Microsoft, in last may launch Cognitive Services(a service for customizing Image recognition models). However, they all are in preview stage as of now. Also, Google has over 10,000 businesses that use Cloud AI Services, which is currently 12% share of the cloud market. Amazon leads the market with 62% as of last quarter of 2017.

Pricing and Access

Google has not officially revealed the pricing information yet, but according to TechCrunch, chances are there may be two categories of fees. One for training the models and the other one for accessing the models through Google’s API. To get access to Alpha Cloud AutoML visions, developers currently have to apply for access.

Guest Blogger (Data Science) at Digital Vidya. A Data passionate who loves reading and diving deeper into the Machine Learning and Data Science arts. Always eager to learn about new research and new ways to solve problems using ML and AI.

  • Data-Analytics

  • Your Comment

    Your email address will not be published.