Troubleshooting - DataSetError: Failed while loading data from data set MlflowMetricsDataSet(prefix=model.train_metrics) Hello everyone! I'm trying to run a job in Databricks to train some models using Kedro and I've come across the following error when saving some model information using mlflow: DataSetError: Failed while loading data from data set MlflowMetricsDataSet(prefix=model.train_metrics). 'RestStore' object has no attribute 'get_all_metrics' Here is the catalog.yml snippet referring to the error:

_base_mlflow_artifact: &base_mlflow_artifact
  type: kedro_mlflow.io.artifacts.MlflowArtifactDataSet

_base_mlflow_metrics: &base_mlflow_metrics
  type: kedro_mlflow.io.metrics.MlflowMetricsDataSet

{{model}}.train_metrics:
  <<: *base_mlflow_metrics

{{model}}.test_metrics:
  <<: *base_mlflow_metrics

Running the pipeline locally on my computer, the error does not occur.

With the caveat that I don't know much about MLFlow and am just reading code... When working with local filesystem,

FileStore

is used; when working with remote tracking server,

RestStore

is used. (Source: https://github.com/mlflow/mlflow/issues/333#issuecomment-415237214)

FileStore

support `get_all_metrics`: https://github.com/mlflow/mlflow/blob/988a74ebb17e79d392cbbc6b2a45ee5f842e6ef6/mlflow/store/tracking/file_store.py#L753

RestStore

does not: https://github.com/mlflow/mlflow/blob/master/mlflow/store/tracking/rest_store.py Similar to this issue, it would probably need to be resolved by somebody implementing the method on `RestStore`: https://github.com/mlflow/mlflow/issues/549 I'd recommend raising an issue on the MLFlow repo with this:

'RestStore' object has no attribute 'get_all_metrics'