LightGBMError: Multiclass objective and metrics don't match

[ansin218]

Does the library have capabilities to handle continuous target variables? I am trying to pass a variable with 600,000 records with integer values between 0-180 and I stumble upon this:

Traceback (most recent call last):

  File "<ipython-input-2-285466e856ff>", line 1, in <module>
    runfile('C:/Users/ankurs/nutella.py', wdir='C:/Users/ankurs')

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 786, in runfile
    execfile(filename, namespace)

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 110, in execfile
    exec(compile(f.read(), filename, 'exec'), namespace)

  File "C:/Users/ankurs/nutella.py", line 89, in <module>
    churnFeatures = TreeFeatureSelection(x, targetChurn)

  File "C:\Users\ankurs\FeatureEngineering.py", line 22, in TreeFeatureSelection
    n_iterations = 10, early_stopping = True)

  File "C:\Users\ankurs\feature_selector.py", line 311, in identify_zero_importance
    early_stopping_rounds = 100, verbose = -1)

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\sklearn.py", line 700, in fit
    callbacks=callbacks)

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\sklearn.py", line 502, in fit
    callbacks=callbacks)

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\engine.py", line 192, in train
    booster = Booster(params=params, train_set=train_set)

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\basic.py", line 1487, in __init__
    train_set.construct().handle,

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\basic.py", line 985, in construct
    categorical_feature=self.categorical_feature, params=self.params)

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\basic.py", line 771, in _lazy_init
    self.__init_from_np2d(data, params_str, ref_dataset)

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\basic.py", line 835, in __init_from_np2d
    ctypes.byref(self.handle)))

  File "C:\Users\ankurs\AppData\Local\Continuum\anaconda3\lib\site-packages\lightgbm\basic.py", line 45, in _safe_call
    raise LightGBMError(decode_string(_LIB.LGBM_GetLastError()))

LightGBMError: Multiclass objective and metrics don't match

I also tried googling and adding multi_logloss eval metrics to the LGBMClassifier and LGBMRegressor but no luck.

[alibaltschun]

@ansin218 did you find the solutions?

[ansin218]

@alibaltschun Unfortunately, not yet.

[alibaltschun]

@ansin218 i found the problem,
coz we installing the up to date packages.

try to downgrade lightgbm
pip install --user lightgbm==2.1.1

if still error try to downgrade seaborn too
pip install --user seaborn==0.8.1

[ansin218]

The first solution itself worked. Thanks a ton, mate!

[alibaltschun]

welcome @ansin218