We’re excited to announce that we’re moving towards the MLflow 1.0 release in the next few weeks. In addition to some exciting new features (much-improved metric visualizations, metric X coordinates, improved search functionality and HDFS support), MLflow 1.0 offers Python, Java, R, and REST API stability.

To that end, we’ve published a release candidate of MLflow 1.0 that you can use to try out all the latest 1.0 features. We’d love to get your feedback and fix any issues that arise before the 1.0 release. Please report issues on GitHub.

Click here for instructions on how to try the latest release candidate.

Thanks for your help testing out 1.0!