class sklearn.feature_extraction.FeatureHasher(n_features=1048576, *, input_type='dict', dtype=<class 'numpy.float64'>, alternate_sign=True)[source]#

Implements feature hashing, aka the hashing trick.

This class turns sequences of symbolic feature names (strings) into scipy.sparse matrices, using a hash function to compute the matrix column corresponding to a name. The hash function employed is the signed 32-bit version of Murmurhash3.

Feature names of type byte string are used as-is. Unicode strings are converted to UTF-8 first, but no Unicode normalization is done. Feature values must be (finite) numbers.

This class is a low-memory alternative to DictVectorizer and CountVectorizer, intended for large-scale (online) learning and situations where memory is tight, e.g. when running prediction code on embedded devices.

For an efficiency comparison of the different feature extractors, see FeatureHasher and DictVectorizer Comparison.

Read more in the User Guide.

Added in version 0.13.

n_featuresint, default=2**20

The number of features (columns) in the output matrices. Small numbers of features are likely to cause hash collisions, but large numbers will cause larger coefficient dimensions in linear learners.

input_typestr, default=’dict’

Choose a string from {‘dict’, ‘pair’, ‘string’}. Either “dict” (the default) to accept dictionaries over (feature_name, value); “pair” to accept pairs of (feature_name, value); or “string” to accept single strings. feature_name should be a string, while value should be a number. In the case of “string”, a value of 1 is implied. The feature_name is hashed to find the appropriate column for the feature. The value’s sign might be flipped in the output (but see non_negative, below).

dtypenumpy dtype, default=np.float64

The type of feature values. Passed to scipy.sparse matrix constructors as the dtype argument. Do not set this to bool, np.boolean or any unsigned integer type.

alternate_signbool, default=True

When True, an alternating sign is added to the features as to approximately conserve the inner product in the hashed space even for small n_features. This approach is similar to sparse random projection.

Changed in version 0.19: alternate_sign replaces the now deprecated non_negative parameter.

See also


Vectorizes string-valued features using a hash table.


Handles nominal/categorical features.


This estimator is stateless and does not need to be fitted. However, we recommend to call fit_transform instead of transform, as parameter validation is only performed in fit.


>>> from sklearn.feature_extraction import FeatureHasher
>>> h = FeatureHasher(n_features=10)
>>> D = [{'dog': 1, 'cat':2, 'elephant':4},{'dog': 2, 'run': 5}]
>>> f = h.transform(D)
>>> f.toarray()
array([[ 0.,  0., -4., -1.,  0.,  0.,  0.,  0.,  0.,  2.],
       [ 0.,  0.,  0., -2., -5.,  0.,  0.,  0.,  0.,  0.]])

With input_type="string", the input must be an iterable over iterables of strings:

>>> h = FeatureHasher(n_features=8, input_type="string")
>>> raw_X = [["dog", "cat", "snake"], ["snake", "dog"], ["cat", "bird"]]
>>> f = h.transform(raw_X)
>>> f.toarray()
array([[ 0.,  0.,  0., -1.,  0., -1.,  0.,  1.],
       [ 0.,  0.,  0., -1.,  0., -1.,  0.,  0.],
       [ 0., -1.,  0.,  0.,  0.,  0.,  0.,  1.]])
fit(X=None, y=None)[source]#

Only validates estimator’s parameters.

This method allows to: (i) validate the estimator’s parameters and (ii) be consistent with the scikit-learn transformer API.


Not used, present here for API consistency by convention.


Not used, present here for API consistency by convention.


FeatureHasher class instance.

fit_transform(X, y=None, **fit_params)[source]#

Fit to data, then transform it.

Fits transformer to X and y with optional parameters fit_params and returns a transformed version of X.

Xarray-like of shape (n_samples, n_features)

Input samples.

yarray-like of shape (n_samples,) or (n_samples, n_outputs), default=None

Target values (None for unsupervised transformations).


Additional fit parameters.

X_newndarray array of shape (n_samples, n_features_new)

Transformed array.


Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.


A MetadataRequest encapsulating routing information.


Get parameters for this estimator.

deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.


Parameter names mapped to their values.

set_output(*, transform=None)[source]#

Set output container.

See Introducing the set_output API for an example on how to use the API.

transform{“default”, “pandas”, “polars”}, default=None

Configure output of transform and fit_transform.

  • "default": Default output format of a transformer

  • "pandas": DataFrame output

  • "polars": Polars output

  • None: Transform configuration is unchanged

Added in version 1.4: "polars" option was added.

selfestimator instance

Estimator instance.


Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.


Estimator parameters.

selfestimator instance

Estimator instance.

set_transform_request(*, raw_X: bool | None | str = '$UNCHANGED$') FeatureHasher[source]#

Request metadata passed to the transform method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

  • True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

  • False: metadata is not requested and the meta-estimator will not pass it to transform.

  • None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

  • str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.


This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

raw_Xstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for raw_X parameter in transform.


The updated object.


Transform a sequence of instances to a scipy.sparse matrix.

raw_Xiterable over iterable over raw features, length = n_samples

Samples. Each sample must be iterable an (e.g., a list or tuple) containing/generating feature names (and optionally values, see the input_type constructor argument) which will be hashed. raw_X need not support the len function, so it can be the result of a generator; n_samples is determined on the fly.

Xsparse matrix of shape (n_samples, n_features)

Feature matrix, for use with estimators or further transformers.