AttributeError: DensityRatioEstimator instance has no attribute '_sigma'

[damhurmuller]

Traceback (most recent call last):
File "challenge.py", line 187, in
weights = kliep.predict(X)
File "challenge.py", line 180, in predict
return np.dot(self._phi(X, sigma=sigma), self._alpha).reshape((X.shape[0],))
File "challenge.py", line 150, in _phi
sigma = self._sigma
AttributeError: DensityRatioEstimator instance has no attribute '_sigma'

[xiaochenwang94]

I have same problem.

[jonmart0304]

I think it needs to be initialized in the constructor first because it is out of scope?

[usccolumbia]

if the runnig does not converge, this code returns without setting the _sigma, so get the error.

        if len(sorted_scores) == 0:   
            warnings.warn('LCV failed to converge for all values of sigma.')
            return self
        self._sigma = sorted_scores[0][0][1]

[ramdhan1989]

I have some problem

[liaochen1988]

I have the same problem

[atakanfilgoz]

As @usccolumbia mentioned if the run does not converge, this code does not set _sigma variable and it throws exception. You can try larger sigmas based on your data to avoid.

[SSMK-wq]

What was done to solve this issue? I also get the same error. Is anyone here tried anything to fix this issue? and how did you fix it? I already tried changing the list of values (for sigma) in code to larger values such as 100