Блог

Я пытаюсь перекрестно проверить некоторые вещи в python, но используя это:

 from sklearn.model_selection import cross_val_score
 

В контексте я создаю следующую модель:

 coef = poly.polyfit(train["X_training"], train["Y_training"], 4)
model = poly.Polynomial(coef)
 

И теперь я пытаюсь перекрестно проверить это таким образом

 from sklearn.model_selection import cross_val_score
cross_val_score(model,test["X_test"], test["Y_test"],cv=5)
 

Но ошибки, которые у меня есть, это:

 ---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-186-5bcb7ab926a4> in <module>
      1 from sklearn.model_selection import cross_val_score
----> 2 cross_val_score(model,test["X_test"], test["Y_test"],cv=0)

~anaconda3envsTensonFlowlibsite-packagessklearnutilsvalidation.py in inner_f(*args, **kwargs)
     70                           FutureWarning)
     71         kwargs.update({k: arg for k, arg in zip(sig.parameters, args)})
---> 72         return f(**kwargs)
     73     return inner_f
     74 

~anaconda3envsTensonFlowlibsite-packagessklearnmodel_selection_validation.py in cross_val_score(estimator, X, y, groups, scoring, cv, n_jobs, verbose, fit_params, pre_dispatch, error_score)
    397     """
    398     # To ensure multimetric format is not supported
--> 399     scorer = check_scoring(estimator, scoring=scoring)
    400 
    401     cv_results = cross_validate(estimator=estimator, X=X, y=y, groups=groups,

~anaconda3envsTensonFlowlibsite-packagessklearnutilsvalidation.py in inner_f(*args, **kwargs)
     70                           FutureWarning)
     71         kwargs.update({k: arg for k, arg in zip(sig.parameters, args)})
---> 72         return f(**kwargs)
     73     return inner_f
     74 

~anaconda3envsTensonFlowlibsite-packagessklearnmetrics_scorer.py in check_scoring(estimator, scoring, allow_none)
    423             return None
    424         else:
--> 425             raise TypeError(
    426                 "If no scoring is specified, the estimator passed should "
    427                 "have a 'score' method. The estimator %r does not."

TypeError: If no scoring is specified, the estimator passed should have a 'score' method. The estimator Polynomial([ 6.0000592 ,  8.02956741, -5.99141415, -3.00869471,  1.99588109], domain=[-1,  1], window=[-1,  1]) does not.
 

Я что-то делаю не так?

Ну, похоже, что способ правильной перекрестной проверки — это с

 from sklearn.model_selection import cross_val_score
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import PolynomialFeatures
test = test.dropna()
poly_features = PolynomialFeatures(degree=grade)
X_poly = poly_features.fit_transform(test)
poly = LinearRegression()
cross_val_score(poly, X_poly, test["Y_test"], cv=5)