Troubleshooting¶

Below is a list of error messages that may occur with potential solutions.

Pandas DataError: No numeric types to aggregate

If one of the values stored with finalize_job() has a non-numeric type (e.g. string).

ValueError in sklearn: Input contains NaN/Inf or value too big for float 32

This happens during creation of the PDF report if there are less than two successful jobs in an iteration. May happen, for example, when testing with a very small number of jobs per iteration and then some of them fail.