MCPcopy Create free account
hub / github.com/scikit-learn/scikit-learn / _check_input

Method _check_input

sklearn/preprocessing/_data.py:3602–3647  ·  view source on GitHub ↗

Validate the input before fit and transform. Parameters ---------- X : array-like of shape (n_samples, n_features) in_fit : bool Whether or not `_check_input` is called from `fit` or other methods, e.g. `predict`, `transform`, etc. c

(self, X, in_fit, check_positive=False, check_shape=False)

Source from the content-addressed store, hash-verified

3600 return lmbda
3601
3602 def _check_input(self, X, in_fit, check_positive=False, check_shape=False):
3603 """Validate the input before fit and transform.
3604
3605 Parameters
3606 ----------
3607 X : array-like of shape (n_samples, n_features)
3608
3609 in_fit : bool
3610 Whether or not `_check_input` is called from `fit` or other
3611 methods, e.g. `predict`, `transform`, etc.
3612
3613 check_positive : bool, default=False
3614 If True, check that all data is positive and non-zero (only if
3615 ``self.method=='box-cox'``).
3616
3617 check_shape : bool, default=False
3618 If True, check that n_features matches the length of self.lambdas_
3619 """
3620 X = validate_data(
3621 self,
3622 X,
3623 ensure_2d=True,
3624 dtype=FLOAT_DTYPES,
3625 force_writeable=True,
3626 copy=self.copy,
3627 ensure_all_finite="allow-nan",
3628 reset=in_fit,
3629 )
3630
3631 with warnings.catch_warnings():
3632 warnings.filterwarnings("ignore", r"All-NaN (slice|axis) encountered")
3633 if check_positive and self.method == "box-cox" and np.nanmin(X) <= 0:
3634 raise ValueError(
3635 "The Box-Cox transformation can only be "
3636 "applied to strictly positive data"
3637 )
3638
3639 if check_shape and not X.shape[1] == len(self.lambdas_):
3640 raise ValueError(
3641 "Input data has a different number of features "
3642 "than fitting data. Should have {n}, data has {m}".format(
3643 n=len(self.lambdas_), m=X.shape[1]
3644 )
3645 )
3646
3647 return X
3648
3649 def __sklearn_tags__(self):
3650 tags = super().__sklearn_tags__()

Callers 2

_fitMethod · 0.95
transformMethod · 0.95

Calls 2

validate_dataFunction · 0.90
formatMethod · 0.80

Tested by

no test coverage detected