ModuleNotFoundError: No module named 'databricks.feature_store.mlflow_model'

[SOlivero]

I'm using runtime DBR 12.2 LTS ML
Getting this error when running. 

import mlflow
logged_model = 'runs:/.../model' #my run id

# Load model as a PyFuncModel.
loaded_model = mlflow.pyfunc.load_model(logged_model)

# Predict on a Pandas DataFrame.
import pandas as pd
loaded_model.predict(pd.DataFrame(data))

[SOlivero]

In mi logged model the MLModel has the following flavor:

flavors:
  python_function:
    data: data/feature_store
    env:
      conda: conda.yaml
      virtualenv: python_env.yaml
    loader_module: databricks.feature_store.mlflow_model
    python_version: 3.9.5

I think here is where the problem is, but I don't know how to solve it 

[Kumaran]

Hi @SOlivero,

Thank you for posting the question in Databricks community.

Please share with us the error message for further analysis.

[hollybaker354]

Hi @SOlivero I'm having the same issue, have you found a solution?

[hollybaker354]

[Kumaran]

Hi @hollybaker354 and @SOlivero,

Thank you for the reply.

Can you check by setting the maxDepth to the default value and let us know how it works?

[Kumaran]

Hi @SOlivero and @hollybaker354,

Can you check by setting the maxDepth to the default value and see how it works?

[x_y]

hi @Kumaran 

Could you please let me know where to set the "maxDepth" to the default value? What do you mean?

I'm having the same issue. Below please see the error, the artifacts of the logged model, and how I logged the model with feature store.

fs.log_model()error messagelog_model artifacts