Google lets developers add machine learning to their apps
Cloud AutoML, Google's machine learning integration tool, will be made available in stages
Google has opened up its Cloud AutoML tool to address the widening skills gap businesses are experiencing in the machine learning and AI space.
Although the company has already released pre-trained machine learning models to developers via APIs, they're still too complicated for businesses without a specific machine learning resource to integrate the tech across the entire business.
So they launched a much simpler product, Cloud AutoML, which offers machine learning capabilities to integrate into applications in less than a day.
"Currently, only a handful of businesses in the world have access to the talent and budgets needed to fully appreciate the advancements of ML and AI," Jia Li, head of R&D and Fei-Fei Li, chief scientist at Google Cloud AI, said in a statement.
"There's a very limited number of people that can create advanced machine learning models. And if you're one of the companies that has access to ML/AI engineers, you still have to manage the time-intensive and complicated process of building your own custom ML model."
Cloud AutoML enables businesses without the expertise to build models using learning2learn and transfer learning from Google.
It will release its Cloud AutoML in stages, with the first feature coming being Cloud AutoML Vision, a resource to help organisations create custom machine learning models for image recognition. With an easy-to use drag and drop UI, developers can upload images, train and manage models, which can be sent straight to Google Cloud.
"Early results using Cloud AutoML Vision to classify popular public datasets like ImageNet and CIFAR have shown more accurate results with fewer misclassifications than generic ML APIs," Li and Li explained.
Picture: Bigstock
