TensorFlow Brings Machine Learning Decision Forests to Google Sheets
Designed for ease of accessibility, Simple ML delivers TensorFlow Yggdrasil Decision Tree machine learning right in Google Sheets.
The TensorFlow team has announced a project to bring machine learning to a broader audience — by integrating a simplified version into Google's Sheets spreadsheet platform, as the add-on Simple ML.
'Simple ML is an add-on, in beta, for Google Sheets from the TensorFlow team that helps make machine learning accessible to all," its creators, members of the TensorFlow Decision Forests team in Google Zurich, claim. "Anyone, even people without programming or ML expertise, can experiment and apply some of the power of machine learning to their data in Google Sheets with just a few clicks. From small business owners, scientists, and students to business analysts at large corporations, anyone familiar with Google Sheets can make valuable predictions automatically."
Tailored specifically for handling data commonly found in spreadsheet format, Simple ML comes with a range of pre-defined tasks — including prediction of values missing in a series or detection of abnormal values. There's no programming or even manual training involved: the user just selects the data, runs Simple ML, and chooses what they want the platform to do.
For those who want to dig deeper, however, it's possible to take over: Models can be manually trained, evaluated, and exported for external use. "Even if you already know how to train and use machine learning models, Simple ML in Sheets can help make your life even easier," the team claims. "For instance, training, evaluating, interpreting, and exporting a model to Notebook takes only five clicks and as little as 10 seconds.
"Since Simple ML in Sheets is based on state-of-the-art ML technology that also powers TensorFlow Decision Forests, and pre-optimized, you might even get better models."
More details on Simple ML for Sheets is available on the project website, along with tutorials on its use; in its initial release, the software exports to TensorFlow, Colab, and TF Serving, along with support for calling the model from C++, Go, and JavaScript.