Creating an Input Schema for Multiple DataFrames in MLflow

Machine Learning

Dive into the world of machine learning on the Databricks platform. Explore discussions on algorithms, model training, deployment, and more. Connect with ML enthusiasts and experts.

Hi everyone,

I am working with MLflow version 2.5.0 and need to create an input_schema for my model. My data schema is divided into several DataFrames, for example:

{
"dataframe_split": {
"columns": ["ClientGuid", "Instance", "TypeScore", ...],
"data": [["26c5f035-cf57-4cb7-b67c-55da528ce3e0", "1", "POST", ...]]
},
"dataframe_loan_split": {
"columns": ["System", "Entity", "Credit", ...],
"data": [["R", "Entity1", "M1 - MIC. IND. DEB. GAR.", ...]]
},
...
}

Could someone provide an example of how to define an input_schema using MLflow for this type of structure? I am looking for a way to specify all the DataFrames within the input schema.

Thanks for your help.