Amazon has released a new open source Python library designed to "truly democratize machine learning" by making it available to as many developers as possible, even without prior experience: AutoGluon.
"We developed AutoGluon to truly democratize machine learning, and make the power of deep learning available to all developers," claims Amazon Web Services' Jonas Mueller of the project. "Due to the inherently opaque nature of deep learning, many of the choices made by deep-learning experts are based on ad hoc intuition, rather than a rigorous scientific understanding of how individual choices affect desired outcomes. AutoGluon solves this problem, as all choices are automatically tuned within default ranges that are known to perform well for the particular task and model."
The result, Amazon claims, is the ability for developers to create a high-performance neural network model in as little as three lines of code — and without the manual experimentation normally associated with tuning the various choices associated with the network. In this latter feature, it's not dissimilar to the precocial neural network approach demonstrated by AWS rival Google Brain earlier this month.
Following internal testing, Amazon has released AutoGluon for public use via an official website boasting tutorials on leveraging deep learning technology on tabular, free-text, and image data across classification, regression, and object detection tasks, along with information on using the AutoGluon APIs within custom applications.
More information on the development of AutoGluon is also available on the Amazon Science website.