ValueError                                Traceback (most recent call last)
[<ipython-input-2-d8e360602655>](https://localhost:8080/#) in <cell line: 20>()
     18 
     19 my_est = MyEstimator()
---> 20 my_est.fit_transform(np.array([5, 7, np.nan, 9]).reshape(2, 2))

7 frames
[/usr/local/lib/python3.10/dist-packages/sklearn/utils/_set_output.py](https://localhost:8080/#) in wrapped(self, X, *args, **kwargs)
    317     @wraps(f)
    318     def wrapped(self, X, *args, **kwargs):
--> 319         data_to_wrap = f(self, X, *args, **kwargs)
    320         if isinstance(data_to_wrap, tuple):
    321             # only wrap the first output for cross decomposition

[/usr/local/lib/python3.10/dist-packages/sklearn/base.py](https://localhost:8080/#) in fit_transform(self, X, y, **fit_params)
    857         if y is None:
    858             # fit method of arity 1 (unsupervised transformation)
--> 859             return self.fit(X, **fit_params).transform(X)
    860         else:
    861             # fit method of arity 2 (supervised transformation)

[/usr/local/lib/python3.10/dist-packages/sklearn/utils/_set_output.py](https://localhost:8080/#) in wrapped(self, X, *args, **kwargs)
    317     @wraps(f)
    318     def wrapped(self, X, *args, **kwargs):
--> 319         data_to_wrap = f(self, X, *args, **kwargs)
    320         if isinstance(data_to_wrap, tuple):
    321             # only wrap the first output for cross decomposition

[/usr/local/lib/python3.10/dist-packages/sklearn/feature_selection/_base.py](https://localhost:8080/#) in transform(self, X)
    105         # note: we use get_tags instead of __sklearn_tags__ because this is a
    106         # public Mixin.
--> 107         X = validate_data(
    108             self,
    109             X,

[/usr/local/lib/python3.10/dist-packages/sklearn/utils/validation.py](https://localhost:8080/#) in validate_data(_estimator, X, y, reset, validate_separately, skip_check_array, **check_params)
   2931             out = X, y
   2932     elif not no_val_X and no_val_y:
-> 2933         out = check_array(X, input_name="X", **check_params)
   2934     elif no_val_X and not no_val_y:
   2935         out = _check_y(y, **check_params)

[/usr/local/lib/python3.10/dist-packages/sklearn/utils/validation.py](https://localhost:8080/#) in check_array(array, accept_sparse, accept_large_sparse, dtype, order, copy, force_writeable, force_all_finite, ensure_all_finite, ensure_non_negative, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, estimator, input_name)
   1104 
   1105         if ensure_all_finite:
-> 1106             _assert_all_finite(
   1107                 array,
   1108                 input_name=input_name,

[/usr/local/lib/python3.10/dist-packages/sklearn/utils/validation.py](https://localhost:8080/#) in _assert_all_finite(X, allow_nan, msg_dtype, estimator_name, input_name)
    118         return
    119 
--> 120     _assert_all_finite_element_wise(
    121         X,
    122         xp=xp,

[/usr/local/lib/python3.10/dist-packages/sklearn/utils/validation.py](https://localhost:8080/#) in _assert_all_finite_element_wise(X, xp, allow_nan, msg_dtype, estimator_name, input_name)
    167                 "#estimators-that-handle-nan-values"
    168             )
--> 169         raise ValueError(msg_err)
    170 
    171 

ValueError: Input X contains NaN.
MyEstimator does not accept missing values encoded as NaN natively. For supervised learning, you might want to consider sklearn.ensemble.HistGradientBoostingClassifier and Regressor which accept missing values encoded as NaNs natively. Alternatively, it is possible to preprocess the data, for instance by using an imputer transformer in a pipeline or drop samples with missing values. See https://scikit-learn.org/stable/modules/impute.html You can find a list of all estimators that handle NaN values at the following page: https://scikit-learn.org/stable/modules/impute.html#estimators-that-handle-nan-values