Which code block will produce a DataFrame with the schema containing "customer_id" as LONG type and "predictions" as DOUBLE type, given that:

A production model has been logged in MLflow by the data science team

The model takes a list of column names as input and outputs a DOUBLE type column

The customers table (with customer_id as key column) is already loaded into a DataFrame

The required feature columns for the model are properly defined?

Exam-Like

df.map(lambda x:model(x[columns])).select("customer_id, predictions")_

12.5%

df.select("customer_id", model(columns).alias("predictions"))_

55.6%

model.predict(df, columns)

6.3%

df.select("customer_id", pandas_udf(model, columns).alias("predictions"))

11.8%

df.apply(model, columns).select("customer_id, predictions")_

13.9%

Databricks Certified Data Engineer - Professional