BigML Release: Automatically Find the Optimal Machine Learning Model with OptiML!

Posted by

BigML’s new release is here! Join us on Wednesday, May 16, 2018, at 10:00 AM PDT (Portland, Oregon. GMT -07:00) / 07:00 PM CEST (Valencia, Spain. GMT +02:00) for a FREE live webinar to discover the latest version of the BigML platform. We will be presenting OptiMLa new BigML resource that automatically finds the best performing supervised model for your data to help you solve classification and regression problems with a single click.

BigML’s mission remains unchanged since its inception: to make Machine Learning easy and beautiful for everyone. As an important milestone in this journey, we are bringing our newest feature, OptiML, to the BigML Dashboard, API, and WhizzML.

Even after you define your Machine Learning problem, collect and pre-process data, and generate relevant features, it can still be very time-consuming and difficult to choose the best algorithm for your problem. For non-experts, this challenge becomes even more pronounced as they have to understand and configure many parameters before they land on an optimal model (often via trial and error), with no indication if they should continue surveying additional options. On the other hand, advanced users also appreciate the time savings resulting from best practice optimization techniques in searching their hypothesis spaces when benchmarking against their own hand-fit models. With BigML, regardless of your previous Machine Learning experience, you can now automatically tune your supervised models to quickly find an optimal model to effectively solve your classification or regression problems. The new OptiML capability listed under our supervised menu on BigML Dashboard enables just that with a single click!

OptiML automatically creates and evaluates multiple supervised models (decision trees, ensembles, logistic regressions, and deepnets) with different configurations by using Bayesian parameter optimization. Put simply, OptiML tries new values for groups of parameters, trains models, evaluates them, and tries a new group of parameters based on the results of the previous trials. When this dynamic process finishes, you will get a list of the models best fitting your data, so you can compare them and select the one that best suits your predictive use case. OptiML is an integration of the SMACdown algorithm into the BigML Dashboard that puts more Machine Learning in your Machine Learning.

For ease of use and completeness sake, in addition to finding the best supervised model among several algorithms with OptiML, we have also enabled the Automatic Optimization option for models, ensembles, logistic regressions, and deepnets, individually. This means that you no longer need to manually tune any of your supervised models to achieve the best results. Instead, you can just select the Automatic Optimization option and BigML will execute an automatic optimization task for your chosen algorithm. Once complete, it will return the top performing model ready to be used.

Want to know more about OptiML?

If you have any questions or you would like to learn more about how OptiML works, please visit the release page. It includes a series of blog posts, the BigML Dashboard and API documentation, the webinar slideshow as well as the full webinar recording.

One comment

  1. Reblogged this on BLACK BOX PARADOX and commented:
    BigML now allows you to finds the best performing supervised model for your data automatically to help you solve classification and regression problems with a single click.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s