In scikit-learn, calibrators_ attribute in the _CalibratedClassifier class has been renamed into calibrators (see https://github.com/scikit-learn/scikit-learn/commit/8471c8389d794309b0b62a353e3af903acd48223). This cased AttributeError: '_CalibratedClassifier' object has no attribute 'calibrators_' when using the CalibratedClassifierModel recipe. This PR renames the usages of calibrators_ attribute to calibrators to fix that error.
scikit-learn,calibrators_attribute in the_CalibratedClassifierclass has been renamed intocalibrators(see https://github.com/scikit-learn/scikit-learn/commit/8471c8389d794309b0b62a353e3af903acd48223). This casedAttributeError: '_CalibratedClassifier' object has no attribute 'calibrators_'when using theCalibratedClassifierModelrecipe. This PR renames the usages ofcalibrators_attribute tocalibratorsto fix that error.ml_insights,SplineCalibclass constructor has changed. Thepenalty,solverparams have been removed andmethodparam has been added. (see https://github.com/numeristical/introspective/commit/9d4f5be0b159269bf61d66a0364b151f9aec5584#diff-6298dc7bfb7d0111ff50669c3514cc651dbfdc201be31e81d1d697df1db9df8f, https://github.com/numeristical/splinecalib/commit/45dc522c000c8e90243ce7d5024bc26b2f211f36#diff-b39b488cbeef80f1d0137a05757ab135c3037ea3129c140eb108b26458c89625). Hence this PR changes theSplineCalibconstructor call accordingly.