'custom metric collection in katib(AutoML) on kubeflow

from the example of pytorch example in Katib, it can collect the metric for loss and accuracy. I tried to calculate other measurements during train() and wanted to be collect to display in katib(AutoML) ui, I already put in logging.info() and also put it in spec.objective.additionalMetricNames in yaml file, but it didn't work.

kubernetespytorchautomlkubeflow


Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source

Related Questions