In 2020, we are expecting that AutoAI users will start to benefit from NeuNetS in a growing number of use cases

Since the beta release of the NeuNetS tool in Decemeber 2018, we have received a lot of positive feedback from the Watson Studio user community. The voices from our users inspire us to evolve NeuNetS into its next phase—AutoAI.

AutoAI is the Watson Studio tool focusing on automating the end-to-end model building lifecycle, from data ingestion to model deployment. By automating tasks that typically take data scientists days or weeks, Watson Studio continues its mission of boosting the productivity of data science teams large and small.

NeuNetS will boost AutoAI use cases

Merging NeuNetS into AutoAI allows for the automation and enhancement of a new suite of neural network models, and it addresses much broader data science use cases, including tabular data prediction and classification, time-series forecasting, text mining, and image recognition.

The merge is in progress, and in 2020, we are expecting that AutoAI users will start to benefit from NeuNetS in a growing number of use cases. Now is a good time to get familiar with AutoAI and we are inviting users to try out the new code generation feature. You are welcome to join our early access!

Phasing out NeuNetS beta in Watson Studio

Related to this announcement, we will be phasing out the current NeuNetS beta model builder tool in Watson Studio on December 6, 2019. NeuNetS models stored and deployed to Watson Machine Learning must be downloaded and migrated to newer versions of Keras models. More info can be found in the Watson Studio documentation.