Is your feature request related to a problem?
DataFrame.iterrows
does not preserve types. There are lots of users reporting issues related to this. Even though the behavior is documented, why do we want to have a method that get used a lot but keep surprising users? I bet most users won't read the doc thoroughly until they get bitten by the issue.DataFrame.iterrows
is much slower than other ways of iterating over rows of a DataFrame anyway.
Describe the solution you'd like
Deprecate DataFrame.iterrows
.
API breaking implications
[this should provide a description of how this feature will affect the API]
DataFrame.iterrows
will be deprecated in future.
Describe alternatives you've considered
Maybe rename iterrows
to a better name as suggested in the issue https://github.com/pandas-dev/pandas/issues/43245?
Additional context
Comment From: jreback
-1 on renaming would be +0 in deprecation as this is a highly used method
Comment From: phofl
-1 on renaming too. Would be 0+ on deprecating to. Maybe adding a DeprecationWarning as we did with empty Series dtypes to gather feedback?
Comment From: jreback
-1 on renaming too. Would be 0+ on deprecating to. Maybe adding a DeprecationWarning as we did with empty Series dtypes to gather feedback?
DeprecationWarning is prob all we can do though not user visible but still useful
Comment From: jbrockmendel
id be fine with deprecating