Bug report: Apply Math Operation doesn't cache and reuse results from previous run

Question

In Azure ML Designer, I have a complex data preprocessing pipeline. One of the main features of Designer pipelines is that they are supposed to cache the results of each run, so that they don't need to be re-executed every time if nothing that they depend on has changed. This feature works correctly for all of the nodes in my graph except for the Apply Math Operation nodes. This results in the second half of my pipeline needing to be re-run every time I add a new node.

Answer 1

Thanks for reaching out to us. We will investigate this deeper and let you know any update.

Regards,
Yutong