Repository navigation
Implement dtypes method #62
Description
I have removed methods DataFrame>>columnTypes and DataSeries>>type and the related functionality in #61 because their implementation was very bad and slowed down all other methods.
However, for exploratory purposes it is still important to know what kind of data is stored in each column.
So we need to think how to implement something like dtype in pandas in the most harmless way:
https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.dtypes.html
Activity
Ducasse commentedon Mar 9, 2019
What was your logic?
Because you can have a simple heuristic and just take the first 5 entries and check their types.
Next question is what were you using them for?
Stef