Error when trying to import matplotlib into IPython notebook

[plantgustaf]

Hi, I'm getting the following error:

import matplotlib
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-1-82be63b7783c> in <module>()
----> 1 import matplotlib

/Users/Gustaf/anaconda/lib/python3.5/site-packages/matplotlib/__init__.py in <module>()
   1129 
   1130 # this is the instance used by the matplotlib classes
-> 1131 rcParams = rc_params()
   1132 
   1133 if rcParams['examples.directory']:

/Users/Gustaf/anaconda/lib/python3.5/site-packages/matplotlib/__init__.py in rc_params(fail_on_error)
    973         return ret
    974 
--> 975     return rc_params_from_file(fname, fail_on_error)
    976 
    977 

/Users/Gustaf/anaconda/lib/python3.5/site-packages/matplotlib/__init__.py in rc_params_from_file(fname, fail_on_error, use_default_template)
   1098         parameters specified in the file. (Useful for updating dicts.)
   1099     """
-> 1100     config_from_file = _rc_params_in_file(fname, fail_on_error)
   1101 
   1102     if not use_default_template:

/Users/Gustaf/anaconda/lib/python3.5/site-packages/matplotlib/__init__.py in _rc_params_in_file(fname, fail_on_error)
   1016     cnt = 0
   1017     rc_temp = {}
-> 1018     with _open_file_or_url(fname) as fd:
   1019         try:
   1020             for line in fd:

/Users/Gustaf/anaconda/lib/python3.5/contextlib.py in __enter__(self)
     57     def __enter__(self):
     58         try:
---> 59             return next(self.gen)
     60         except StopIteration:
     61             raise RuntimeError("generator didn't yield") from None

/Users/Gustaf/anaconda/lib/python3.5/site-packages/matplotlib/__init__.py in _open_file_or_url(fname)
    998     else:
    999         fname = os.path.expanduser(fname)
-> 1000         encoding = locale.getdefaultlocale()[1]
   1001         if encoding is None:
   1002             encoding = "utf-8"

/Users/Gustaf/anaconda/lib/python3.5/locale.py in getdefaultlocale(envvars)
    557     else:
    558         localename = 'C'
--> 559     return _parse_localename(localename)
    560 
    561 

/Users/Gustaf/anaconda/lib/python3.5/locale.py in _parse_localename(localename)
    485     elif code == 'C':
    486         return None, None
--> 487     raise ValueError('unknown locale: %s' % localename)
    488 
    489 def _build_localename(localetuple):

ValueError: unknown locale: UTF-8

Any ideas how to fix this? Thanks!

[jenshnielsen]

I think this is a duplicate of #5420 which is really a OSX bug.

The work around is to set LANG and possiblly export LC_ALL to valid values such as some thing like the following:

export LANG=en_US.UTF-8
export LC_ALL=en_US.UTF-8

[jenshnielsen]

I updated the formatting of your output to make it more readable

[plantgustaf]

Thanks so much! That worked!

[plantgustaf]

Do I have to enter this every time I start the IPython notebook? Because it seems like that I have to do this.

[jenshnielsen]

The best solution is to add it to your shell customisation so it's run every time you start a shell. You are probably running bash so it should be either .bash_profile or .bashrc in your home folder

[jenshnielsen]

Could I get you to try the following to review #5931 which is a probably better work around for this issue.
In the notebook where importing matploltlib fails what does do

import locale
locale.getpreferredencoding(do_setlocale=False)

see #5931 which may be the right fix for this.

[plantgustaf]

Ok the fix for me:
Enter in Terminal: touch ~/.bash_profile; open ~/.bash_profile
Add

export LANG=en_US.UTF-8
export LC_ALL=en_US.UTF-8

to the end of the file.
Save & Quit.

[samanemt]

Hi, I'm getting the following error:
ValueError Traceback (most recent call last)
in ()
----> 1 mlp.fit(X_train,y_train)

C:\Users\samaneh\Anaconda3\lib\site-packages\sklearn\neural_network\multilayer_perceptron.py in fit(self, X, y)
616 self : returns a trained MLP model.
617 """
--> 618 return self._fit(X, y, incremental=False)
619
620 @Property

C:\Users\samaneh\Anaconda3\lib\site-packages\sklearn\neural_network\multilayer_perceptron.py in _fit(self, X, y, incremental)
328 hidden_layer_sizes)
329
--> 330 X, y = self._validate_input(X, y, incremental)
331 n_samples, n_features = X.shape
332

C:\Users\samaneh\Anaconda3\lib\site-packages\sklearn\neural_network\multilayer_perceptron.py in _validate_input(self, X, y, incremental)
906 if not incremental:
907 self._label_binarizer = LabelBinarizer()
--> 908 self.label_binarizer.fit(y)
909 self.classes = self.label_binarizer.classes
910 else:

C:\Users\samaneh\Anaconda3\lib\site-packages\sklearn\preprocessing\label.py in fit(self, y)
296 self.y_type_ = type_of_target(y)
297 if 'multioutput' in self.y_type_:
--> 298 raise ValueError("Multioutput target data is not supported with "
299 "label binarization")
300 if _num_samples(y) == 0:

ValueError: Multioutput target data is not supported with label binarization
Any ideas how to fix this? Thanks!

[efiring]

@samanemt, the traceback shows that the error is coming from scikit-learn, not from matplotlib.