DOC: add an example to PeriodIndex.to_timestamp to clarify the behavior in case len(index)<3
#57384
Conversation
|
@MarcoGorelli, I added an example to clarify the behavior of |
pandas/core/arrays/period.py
Outdated
| @@ -636,6 +636,12 @@ def to_timestamp(self, freq=None, how: str = "start") -> DatetimeArray: | |||
| >>> idx.to_timestamp() | |||
| DatetimeIndex(['2023-01-01', '2023-02-01', '2023-03-01'], | |||
| dtype='datetime64[ns]', freq='MS') | |||
|
|
|||
| We can not infer the frequency if len(index) < 3: | |||
There was a problem hiding this comment.
No big deal, but I'd personally phrase this as The frequency will not be inferred if the index contains less than three elements:. I had to read the current version couple of times to understand it.
There was a problem hiding this comment.
thanks for the comment @datapythonista, I corrected docstring as you suggested. I also added a new example to emphasize importance values of index to be strictly monotonic.
pandas/core/arrays/period.py
Outdated
| The frequency will not be inferred if the index contains less than | ||
| three elements and values of index are not strictly monotonic: |
There was a problem hiding this comment.
the way this is phrased it sounds like (pseudocode):
if len(index) < 3 and not index.is_monotonic():
dont_infer_frequency(...)
Instead, I think it's more like
if len(index) < 3 or not index.is_monotonic():
dont_infer_frequency(...)
?
How about
The frequency will not be inferred if the index contains less than
three elements, or if the values of index are not strictly monotonic:
There was a problem hiding this comment.
thanks, you are right, my wording isn't correct. I changed it as you suggested.
…vior in case len(index)<3 (pandas-dev#57384) * add an example to PeriodIndex.to_timestamp * correct to_timestamp docs, add an example * correct the wording in to_timestamp docstring
xref #56213
added an example to
PeriodIndex.to_timestampto clarify the behavior in caselen(index) < 3