Check table for invalid values when fitting #655
Description
Is your feature request related to a problem?
When fitting a model, passing a table containing missing values or non-numerical data will result in either a numpy error or in all future predictions of the model being NaN.
Desired solution
In the InputConversionLayer, check the passed table for invalid data and throw a clear error if so.
Possible alternatives (optional)
No response
Screenshots (optional)
No response
Additional Context (optional)
No response