Confusing error message in OrdinalEncoder when passing single list of categories #13446
Description
Small example:
In [38]: from sklearn.preprocessing import OrdinalEncoder
In [39]: X = np.array([['L', 'M', 'S', 'M', 'L']], dtype=object).T
In [40]: ohe = OrdinalEncoder(categories=['S', 'M', 'L'])
In [41]: ohe.fit(X)
...
ValueError: Shape mismatch: if n_values is an array, it has to be of shape (n_features,).The error message is still using the old n_values, which makes it very confusing.
(another issue is that we might be able to actually detect this case)
Versions
System:
python: 3.7.1 | packaged by conda-forge | (default, Feb 18 2019, 01:42:00) [GCC 7.3.0]
executable: /home/joris/miniconda3/bin/python
machine: Linux-4.4.0-142-generic-x86_64-with-debian-stretch-sid
BLAS:
macros: HAVE_CBLAS=None
lib_dirs: /home/joris/miniconda3/lib
cblas_libs: openblas, openblas
Python deps:
pip: 19.0.2
setuptools: 40.8.0
sklearn: 0.20.2
numpy: 1.16.1
scipy: 1.2.1
Cython: None
pandas: 0.23.4