Profile differences weren't being computed for data_type_stats profiles. I discovered this because psi was only being computed for the categorical profiler. Simply am adding in a line to compute that.
Additionally, dataprofiler/profilers/text_column_profile.py was using the numeric stats mix in to compute diff simply to double check that the data type was the same. I've updated that to use the BaseColumnProfiler diff instead.
Details
Profile differences weren't being computed for
data_type_stats
profiles. I discovered this because psi was only being computed for the categorical profiler. Simply am adding in a line to compute that.Additionally,
dataprofiler/profilers/text_column_profile.py
was using the numeric stats mix in to compute diff simply to double check that the data type was the same. I've updated that to use theBaseColumnProfiler
diff instead.