> > NB. There are varied amounts of consultants so splitting across columns is > > uneven. if it was even melt seems like it would be good > > https://dfrieds.com/data-analysis/melt-unpivot-python-pandas > > Keep it simple: https://docs.python.org/3.6/library/string.html > > -- > Regards =dn
I have managed to split but not filter the numbers. All other code being the same. ... completed_tasks = df.loc[(df['Completed'] == 'Yes') & (df['Motor/Property'] == 'Motor') & (df['Delivery Method'] == 'Group Coaching')] completed_tasks['Consultant'] = completed_tasks['Consultant'].str.split(pat = ";") pattern = "\\#d" completed_tasks['Consultant'] = completed_tasks['Consultant'].str.split(pat = ";") completed_tasks['Consultant'] = completed_tasks['Consultant'].str.contains(pattern) ... Works without the regex which causes this error AttributeError: Can only use .str accessor with string values, which use np.object_ dtype pattern = "\\#d" completed_tasks['Consultant'] = completed_tasks['Consultant'].str.split(pat = ";") Sayth -- https://mail.python.org/mailman/listinfo/python-list
