Hi, I'm trying to use the AutoML API to execute an AutoML experiment using sql server data which contains double-precision floating point numbers. Using the AutoML UI it works seamlessly so I know it can be done. I have tried using the preFeaturizer with a bunch of transforms [mlContext.Transforms.Conversion.ConvertType(name, null, DataKind.Single] to convert the double-precision floating point numbers to single-precision, but I still keep getting the error when I try to call the .Execute() method even DESPITE passing in these preFeaturizer transforms: "Only supported feature column types are Boolean, Single, and String". Is there some sort of trick to this or is it a bug??
This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)