This package uses the underlying data from the program "gender" by Jorg Michael (described here). Its use is pretty straightforward:
>>> import gender_guesser.detector as gender >>> d = gender.Detector() >>> print(d.get_gender(u"Bob")) male >>> print(d.get_gender(u"Sally")) female >>> print(d.get_gender(u"Pauley")) # should be androgynous andy
The result will be one of unknown
(name not found), andy
(androgynous), male
, female
, mostly_male
, or mostly_female
. The difference between andy
and unknown
is that the former is found to have the same probability to be male than to be female, while the later means that the name wasn't found in the database.
I18N is fully supported:
>>> print(d.get_gender(u"\xc1lfr\xfan")) # u"Álfrún" female
Additionally, you can give preference to specific countries:
>>> print(d.get_gender(u"Jamie")) mostly_female >>> print(d.get_gender(u"Jamie", u'great_britain')) mostly_male
Additionally, you can create a detector that is not case sensitive (default is to be case sensitive):
>>> d = gender.Detector(case_sensitive=False) >>> print(d.get_gender(u"sally")) female >>> print(d.get_gender(u"Sally")) female
Try to avoid creating many Detectors, as each creation means reading the data file.
The generator code is distributed under the GPLv3. The data file nam_dict.txt is released under the GNU Free Documentation License.
- Remove
unknown_value
init option, since it can be implemented very easily with a wrapper if needed. - Return
unknown
when name is not found andandy
when it is valid equally for both male and female. - Test README examples as doctests.
- Fix incorrect country-wise gender detection for non-iso886-15 names coming from line length change after data file conversion to UTF-8. See #gh2. Thanks @miquelcamprodon.
- Wire in
tox
to test in both Python 2 and Python 3. - Python 2 and 3 compatiblity.
- Remove obsolete character mapper code.
For previous versions, see sexmachine.
This is a fork of the SexMachine
package by Ferhat Elmas. It was created to be able to publish a Python 3 compatible version to PyPI and to be able add some more improvements without bugging the original author.