Data fetch from Database for TransmogrifAI

Question

Could you please someone help me for fetching data directly from any MySQL Database instead of using any dataset for TransmogrifAI. If possible can I get the code regarding this or any reference?

Answer 1

Here is a guide on how you can connect to MySQL from Apache Spark.

Once you have you Dataset / RDD materialized you can plug it into TransmogrifAI:

1. Either by automatically materializing features as shown here .
2. Or you can create features manually with FeatureBuilder.fromRow function. Then construct OpWorkflow and pass the Dataset into it: workflow. setInputDataset(data) workflow. setInputDataset(data) .