Use format string: '%d%b%y:%H:%M:%S' and pass this as the format for to_datetime, you can find the format options in the docs:
'%d%b%y:%H:%M:%S'
to_datetime
In [73]:pd.to_datetime('04SEP12:00:00:00', format='%d%b%y:%H:%M:%S')Out[73]:Timestamp('2012-09-04 00:00:00')
Another way is simply do:
pd.to_datetime("May 15, 2016", infer_datetime_format=True)>>>Timestamp('2016-05-15 00:00:00')
As another example,
pd.to_datetime("May 15, 2016", format="%b %d, %Y")>>>Timestamp('2016-05-15 00:00:00')
Use %b for the month names.