Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Oftentimes, we are facing a situation where the pipeline for training includes steps for the label, which is not present during prediction, leading to exceptions. Having a way to apply preprocessing steps based on conditions (perhaps through a wrapper) would be very useful.
Use case
This was an issue that became apparent during the AutoML hackathon project.
Description
Oftentimes, we are facing a situation where the pipeline for training includes steps for the label, which is not present during prediction, leading to exceptions. Having a way to apply preprocessing steps based on conditions (perhaps through a wrapper) would be very useful.
Use case
This was an issue that became apparent during the AutoML hackathon project.