model-deployment

I have a simple regression task (using a LightGBMRegressor) where I want to penalize negative predictions more than positive ones. Is there a way to achieve this with the default regression LightGBM objectives (see https://lightgbm.readthedocs.io/en/latest/Parameters.html)? If not, is it somehow possible to define (many example for default LightGBM model) and pass a custom regression objective?

Currently deployment creation allow users to set an env var and pass to the BentoServer container. However in production scenarios, users may want to get env var values directly from existing configmap or secret resources in their k8s cluster. e.g.:

env:
        # Define the environment variable
        - name: SPECIAL_LEVEL_KEY
          valueFrom:
            configMapKeyRef: