Moved sparsefuncs to sparsefuncs_fast

sklearn/cluster/_k_means.pyx

@@ -15,7 +15,7 @@ cimport numpy as np
 cimport cython
 
 from ..utils.extmath import norm
-from sklearn.utils.sparsefuncs cimport add_row_csr
+from sklearn.utils.sparsefuncs_fast cimport add_row_csr
 
 ctypedef np.float64_t DOUBLE
 ctypedef np.int32_t INT

sklearn/cluster/k_means_.py

@@ -19,7 +19,8 @@
 from ..base import BaseEstimator, ClusterMixin, TransformerMixin
 from ..metrics.pairwise import euclidean_distances
 from ..utils.extmath import row_norms
-from ..utils.sparsefuncs import assign_rows_csr, mean_variance_axis0
+from ..utils.sparsefuncs_fast import assign_rows_csr
+from ..utils.sparsefuncs import mean_variance_axis0
 from ..utils import check_arrays
 from ..utils import check_random_state
 from ..utils import atleast2d_or_csr

sklearn/feature_selection/variance_threshold.py

@@ -5,7 +5,7 @@
 from ..base import BaseEstimator
 from .base import SelectorMixin
 from ..utils import atleast2d_or_csr
-from ..utils.sparsefuncs import csr_mean_variance_axis0
+from ..utils.sparsefuncs_fast import csr_mean_variance_axis0
 
 
 class VarianceThreshold(BaseEstimator, SelectorMixin):

sklearn/preprocessing/data.py

@@ -20,9 +20,9 @@
 from ..utils import safe_asarray
 from ..utils import warn_if_not_float
 from ..utils.extmath import row_norms
-from ..utils.sparsefuncs import inplace_csr_row_normalize_l1
-from ..utils.sparsefuncs import inplace_csr_row_normalize_l2
-from ..utils.sparsefuncs import inplace_csr_column_scale
+from ..utils.sparsefuncs_fast import inplace_csr_row_normalize_l1
+from ..utils.sparsefuncs_fast import inplace_csr_row_normalize_l2
+from ..utils.sparsefuncs_fast import inplace_csr_column_scale
 from ..utils.sparsefuncs import mean_variance_axis0
 
 zip = six.moves.zip

sklearn/utils/extmath.py

@@ -17,7 +17,7 @@
 from .fixes import np_version
 from ._logistic_sigmoid import _log_logistic_sigmoid
 from ..externals.six.moves import xrange
-from .sparsefuncs import csr_row_norms
+from .sparsefuncs_fast import csr_row_norms
 from .validation import array2d, NonBLASDotWarning
 
 

sklearn/utils/setup.py

@@ -22,7 +22,7 @@ def configuration(parent_package='', top_path=None):
  libraries.append('m')
  cblas_libs.append('m')
 
- config.add_extension('sparsefuncs', sources=['sparsefuncs.c'],
+ config.add_extension('sparsefuncs_fast', sources=['sparsefuncs_fast.c'],
  libraries=libraries)
 
  config.add_extension('arrayfuncs',

sklearn/utils/sparsefuncs.py

@@ -0,0 +1,58 @@
+# Authors: Manoj Kumar
+
+# License: BSD 3 clause
+import scipy.sparse as sp
+
+from .sparsefuncs_fast import (csr_mean_variance_axis0,
+ csc_mean_variance_axis0,
+ inplace_csr_column_scale,
+ inplace_csc_column_scale) 
+
+def mean_variance_axis0(X):
+ """Compute mean and variance along axis 0 on a CSR or CSC matrix
+
+ Parameters
+ ----------
+ X: CSR or CSC sparse matrix, shape (n_samples, n_features)
+ Input data.
+
+ Returns
+ -------
+
+ means: float array with shape (n_features,)
+ Feature-wise means
+
+ variances: float array with shape (n_features,)
+ Feature-wise variances
+
+ """
+ if isinstance(X, sp.csr_matrix):
+ return csr_mean_variance_axis0(X)
+ elif isinstance(X, sp.csc_matrix):
+ return csc_mean_variance_axis0(X)
+ else:
+ raise TypeError(
+ "Unsupported type; expected a CSR or CSC sparse matrix.")
+
+
+def inplace_column_scale(X, scale):
+ """Inplace column scaling of a CSC/CSR matrix.
+
+ Scale each feature of the data matrix by multiplying with specific scale
+ provided by the caller assuming a (n_samples, n_features) shape.
+
+ Parameters
+ ----------
+ X: CSC or CSR matrix with shape (n_samples, n_features)
+ Matrix to normalize using the variance of the features.
+
+ scale: float array with shape (n_features,)
+ Array of precomputed feature-wise values to use for scaling.
+ """
+ if isinstance(X, sp.csr_matrix):
+ return inplace_csr_column_scale(X, scale)
+ elif isinstance(X, sp.csc_matrix):
+ return inplace_csc_column_scale(X, scale)
+ else:
+ raise TypeError(
+ "Unsupported type; expected a CSR or CSC sparse matrix.")

sklearn/utils/sparsefuncs.pyx → sklearn/utils/sparsefuncs_fast.pyx

@@ -291,56 +291,6 @@ def inplace_csc_column_scale(X, np.ndarray[DOUBLE, ndim=1] scale):
  X_data[j] *= scale[i]
 
 
-def inplace_column_scale(X, np.ndarray[DOUBLE, ndim=1] scale):
- """Inplace column scaling of a CSC/CSR matrix.
-
- Scale each feature of the data matrix by multiplying with specific scale
- provided by the caller assuming a (n_samples, n_features) shape.
-
- Parameters
- ----------
- X: CSC or CSR matrix with shape (n_samples, n_features)
- Matrix to normalize using the variance of the features.
-
- scale: float array with shape (n_features,)
- Array of precomputed feature-wise values to use for scaling.
- """
- if isinstance(X, sp.csr_matrix):
- return inplace_csr_column_scale(X, scale)
- elif isinstance(X, sp.csc_matrix):
- return inplace_csc_column_scale(X, scale)
- else:
- raise TypeError(
- "Unsupported type; expected a CSR or CSC sparse matrix.")
-
-
-def mean_variance_axis0(X):
- """Compute mean and variance along axis 0 on a CSR or CSC matrix
-
- Parameters
- ----------
- X: CSR or CSC sparse matrix, shape (n_samples, n_features)
- Input data.
-
- Returns
- -------
-
- means: float array with shape (n_features,)
- Feature-wise means
-
- variances: float array with shape (n_features,)
- Feature-wise variances
-
- """
- if isinstance(X, sp.csr_matrix):
- return csr_mean_variance_axis0(X)
- elif isinstance(X, sp.csc_matrix):
- return csc_mean_variance_axis0(X)
- else:
- raise TypeError(
- "Unsupported type; expected a CSR or CSC sparse matrix.")
-
-
 @cython.boundscheck(False)
 @cython.wraparound(False)
 cdef void add_row_csr(np.ndarray[np.float64_t, ndim=1] data,

sklearn/utils/tests/test_sparsefuncs.py

@@ -3,9 +3,10 @@
 from numpy.testing import assert_array_almost_equal, assert_array_equal
 
 from sklearn.datasets import make_classification
-from sklearn.utils.sparsefuncs import (assign_rows_csr, mean_variance_axis0,
- inplace_csc_column_scale,
- inplace_csr_column_scale)
+from sklearn.utils.sparsefuncs import mean_variance_axis0
+from sklearn.utils.sparsefuncs_fast import (assign_rows_csr,
+ inplace_csc_column_scale,
+ inplace_csr_column_scale)
 
 def test_mean_variance_axis0():
  X, _ = make_classification(5, 4, random_state=0)