Pandas и scikit-learn: ValueError: Вход содержит NaN, бесконечность или значение, слишком большое для dtype ('float64')

Question

Я пытаюсь приспособить модель Knearest Neigbors к моим данным. но я получаю эту ошибку:

ValueError: Input contains NaN, infinity or a value too large for dtype('float64').

Вот код моего алгоритма Knearest Neigbhors:

def knn_train_test_new(training_col, target_col, df):
    np.random.seed(1)
    df = df.loc[np.random.permutation(len(df))]
#    shuffled_index = np.random.permutation(df.index)
#   df = df.reindex(shuffled_index)

    train_df = df.iloc[0:150] #training set has 75% of the data
    test_df = df.iloc[150:] #test set has 25% of the data  
    k = [5]

    rmse = {}

    for k_val in k:

        model = KNeighborsRegressor(n_neighbors = k_val)
        model.fit(train_df[training_col], train_df[target_col])

        predictions = model.predict(test_df[training_col])

        mse = mean_squared_error(test_df[target_col], predictions)

        rmse[k_val] = (mse ** 0.5)

    return rmse

two_features = ["width", "wheel-base"]
rmse_val = knn_train_test(two_features, 'price', numeric_cars)

И первые пять строк моего кадра данных:

numeric_cars.head()

Я не получаю эту ошибку, когда использую shuffled_index (который я прокомментировал) вместо np.random.permutation. Я не совсем понимаю разницу между ними.

Полный трек ошибок:

ValueError                                Traceback (most recent call last)
<ipython-input-13-03e0dd7acfd0> in <module>()
     25 
     26 two_features = ["width", "wheel-base"]
---> 27 rmse_val = knn_train_test_new(two_features, 'price', numeric_cars)
     28 
     29 #rmse_results = {}

<ipython-input-13-03e0dd7acfd0> in knn_train_test_new(training_col, target_col, df)
     14 
     15         model = KNeighborsRegressor(n_neighbors = k_val)
---> 16         model.fit(train_df[training_col], train_df[target_col])
     17 
     18         predictions = model.predict(test_df[training_col])

~\Anaconda3\lib\site-packages\sklearn\neighbors\base.py in fit(self, X, y)
    743         """
    744         if not isinstance(X, (KDTree, BallTree)):
--> 745             X, y = check_X_y(X, y, "csr", multi_output=True)
    746         self._y = y
    747         return self._fit(X)

~\Anaconda3\lib\site-packages\sklearn\utils\validation.py in check_X_y(X, y, accept_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, multi_output, ensure_min_samples, ensure_min_features, y_numeric, warn_on_dtype, estimator)
    571     X = check_array(X, accept_sparse, dtype, order, copy, force_all_finite,
    572                     ensure_2d, allow_nd, ensure_min_samples,
--> 573                     ensure_min_features, warn_on_dtype, estimator)
    574     if multi_output:
    575         y = check_array(y, 'csr', force_all_finite=True, ensure_2d=False,

~\Anaconda3\lib\site-packages\sklearn\utils\validation.py in check_array(array, accept_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, warn_on_dtype, estimator)
    451                              % (array.ndim, estimator_name))
    452         if force_all_finite:
--> 453             _assert_all_finite(array)
    454 
    455     shape_repr = _shape_repr(array.shape)

~\Anaconda3\lib\site-packages\sklearn\utils\validation.py in _assert_all_finite(X)
     42             and not np.isfinite(X).all()):
     43         raise ValueError("Input contains NaN, infinity"
---> 44                          " or a value too large for %r." % X.dtype)
     45 
     46 

ValueError: Input contains NaN, infinity or a value too large for dtype('float64').