Utilities for Developers¶
Scikit-learn contains a number of utilities to help with development. These are
sklearn.utils, and include tools in a number of categories.
All the following functions and classes are in the module
These utilities are meant to be used internally within the scikit-learn package. They are not guaranteed to be stable between versions of scikit-learn. Backports, in particular, will be removed as the scikit-learn dependencies evolve.
These are tools used to check and validate input. When you write a function which accepts arrays, matrices, or sparse matrices as arguments, the following should be used when applicable.
assert_all_finite: Throw an error if array contains NaNs or Infs.
as_float_array: convert input to an array of floats. If a sparse matrix is passed, a sparse matrix will be returned.
check_array: check that input is a 2D array, raise error on sparse matrices. Allowed sparse matrix formats can be given optionally, as well as allowing 1D or N-dimensional arrays. Calls
check_X_y: check that X and y have consistent length, calls check_array on X, and column_or_1d on y. For multilabel classification or multitarget regression, specify multi_output=True, in which case check_array will be called on y.
indexable: check that all input arrays have consistent length and can be sliced or indexed using safe_index. This is used to validate input for cross-validation.
validation.check_memorychecks that input is
joblib.Memory-like, which means that it can be converted into a
sklearn.utils.Memoryinstance (typically a str denoting the
cachedir) or has the same interface.
If your code relies on a random number generator, it should never use
approach can lead to repeatability issues in unit tests. Instead, a
numpy.random.RandomState object should be used, which is built from
random_state argument passed to the class or function. The function
check_random_state, below, can then be used to create a random
number generator object.
check_random_state: create a
np.random.RandomStateobject from a parameter
np.random, then a randomly-initialized
RandomStateobject is returned.
random_stateis an integer, then it is used to seed a new
RandomStateobject, then it is passed through.
>>> from sklearn.utils import check_random_state >>> random_state = 0 >>> random_state = check_random_state(random_state) >>> random_state.rand(4) array([0.5488135 , 0.71518937, 0.60276338, 0.54488318])
When developing your own scikit-learn compatible estimator, the following helpers are available.
validation.check_is_fitted: check that the estimator has been fitted before calling
predict, or similar methods. This helper allows to raise a standardized error message across estimator.
validation.has_fit_parameter: check that a given parameter is supported in the
fitmethod of a given estimator.
Efficient Linear Algebra & Array Operations¶
extmath.randomized_svd: compute the k-truncated randomized SVD. This algorithm finds the exact truncated singular values decomposition using randomization to speed up the computations. It is particularly fast on large matrices on which you wish to extract only a small number of components.
arrayfuncs.cholesky_delete: (used in
lars_path) Remove an item from a cholesky factorization.
arrayfuncs.min_pos: (used in
sklearn.linear_model.least_angle) Find the minimum of the positive values within an array.
extmath.fast_logdet: efficiently compute the log of the determinant of a matrix.
extmath.density: efficiently compute the density of a sparse vector
extmath.safe_sparse_dot: dot product which will correctly handle
scipy.sparseinputs. If the inputs are dense, it is equivalent to
extmath.weighted_mode: an extension of
scipy.stats.modewhich allows each item to have a real-valued weight.
Efficient Random Sampling¶
random.sample_without_replacement: implements efficient algorithms for sampling
n_samplesintegers from a population of size
Efficient Routines for Sparse Matrices¶
sklearn.utils.sparsefuncs cython module hosts compiled extensions to
sparsefuncs.inplace_csr_column_scale: can be used to multiply the columns of a CSR matrix by a constant scale (one scale per column). Used for scaling features to unit standard deviation in
graph.single_source_shortest_path_length: (not currently used in scikit-learn) Return the shortest path from a single source to all connected nodes on a graph. Code is adapted from networkx. If this is ever needed again, it would be far faster to use a single iteration of Dijkstra’s algorithm from
all_estimators: returns a list of all estimators in scikit-learn to test for consistent behavior and interfaces.
Multiclass and multilabel utility function¶
gen_batches: generator to create slices containing batch size elements from 0 to
safe_mask: Helper function to convert a mask to the format expected by the numpy array or scipy sparse matrix on which to use it (sparse matrices support integer indices only while numpy arrays support both boolean masks and integer indices).
safe_sqr: Helper function for unified squaring (
**2) of array-likes, matrices and sparse matrices.
murmurhash3_32provides a python wrapper for the
MurmurHash3_x86_32C++ non cryptographic hash function. This hash function is suitable for implementing lookup tables, Bloom filters, Count Min Sketch, feature hashing and implicitly defined sparse random projections:
>>> from sklearn.utils import murmurhash3_32 >>> murmurhash3_32("some feature", seed=0) == -384616559 True >>> murmurhash3_32("some feature", seed=0, positive=True) == 3910350737 True
sklearn.utils.murmurhashmodule can also be “cimported” from other cython modules so as to benefit from the high performance of MurmurHash while skipping the overhead of the Python interpreter.