ENH move utility function into dedicated file, not __init__.py

Author: amueller

sklearn/utils/__init__.py

@@ -11,10 +11,12 @@
                          assert_all_finite, array2d, atleast2d_or_csc,
                          atleast2d_or_csr, warn_if_not_float,
                          check_random_state)
+from class_weight import compute_class_weight
 
 __all__ = ["murmurhash3_32", "as_float_array", "check_arrays", "safe_asarray",
            "assert_all_finite", "array2d", "atleast2d_or_csc",
-           "atleast2d_or_csr", "warn_if_not_float", "check_random_state"]
+           "atleast2d_or_csr", "warn_if_not_float", "check_random_state",
+           "compute_class_weight"]
 
 # Make sure that DeprecationWarning get printed
 warnings.simplefilter("always", DeprecationWarning)
@@ -346,52 +348,3 @@ def gen_even_slices(n, n_packs):
 
 class ConvergenceWarning(Warning):
     "Custom warning to capture convergence problems"
-
-
-def compute_class_weight(class_weight, classes, y):
-    """Estimate class weights for unbalanced datasets.
-
-    Parameters
-    ----------
-    class_weight : dict, 'auto' or None
-        If 'auto', class weights will be given inverse proportional
-        to the frequency of the class in the data.
-        If a dictionary is given, keys are classes and values
-        are corresponding class weights.
-        If None is given, the class weights will be uniform.
-    classes : list
-        List of the classes occuring in the data, as given by
-        ``np.unique(y_org)`` with ``y_org`` the original class labels.
-    y : array-like, shape=(n_samples,), dtype=int
-        Array of class indices per sample;
-        0 <= y[i] < n_classes for i in range(n_samples).
-
-
-    Returns
-    -------
-    class_weight_vect : ndarray, shape=(n_classes,)
-        Array with class_weight_vect[i] the weight for i-th class
-        (as determined by sorting).
-    """
-    if class_weight is None or len(class_weight) == 0:
-        # uniform class weights
-        weight = np.ones(classes.shape[0], dtype=np.float64, order='C')
-    elif class_weight == 'auto':
-        # proportional to the number of samples in the class
-        weight = np.array([1.0 / np.sum(y == i) for i in classes],
-                          dtype=np.float64, order='C')
-        weight *= classes.shape[0] / np.sum(weight)
-    else:
-        # user-defined dictionary
-        weight = np.ones(classes.shape[0], dtype=np.float64, order='C')
-        if not isinstance(class_weight, dict):
-            raise ValueError("class_weight must be dict, 'auto', or None,"
-                             " got: %r" % class_weight)
-        for c in class_weight:
-            i = np.searchsorted(classes, c)
-            if classes[i] != c:
-                raise ValueError("Class label %d not present." % c)
-            else:
-                weight[i] = class_weight[c]
-
-    return weight

sklearn/utils/class_weight.py

@@ -0,0 +1,53 @@
+# Authors: Andreas Mueller
+# License: Simplified BSD
+
+import numpy as np
+
+
+def compute_class_weight(class_weight, classes, y):
+    """Estimate class weights for unbalanced datasets.
+
+    Parameters
+    ----------
+    class_weight : dict, 'auto' or None
+        If 'auto', class weights will be given inverse proportional
+        to the frequency of the class in the data.
+        If a dictionary is given, keys are classes and values
+        are corresponding class weights.
+        If None is given, the class weights will be uniform.
+    classes : list
+        List of the classes occuring in the data, as given by
+        ``np.unique(y_org)`` with ``y_org`` the original class labels.
+    y : array-like, shape=(n_samples,), dtype=int
+        Array of class indices per sample;
+        0 <= y[i] < n_classes for i in range(n_samples).
+
+
+    Returns
+    -------
+    class_weight_vect : ndarray, shape=(n_classes,)
+        Array with class_weight_vect[i] the weight for i-th class
+        (as determined by sorting).
+    """
+    if class_weight is None or len(class_weight) == 0:
+        # uniform class weights
+        weight = np.ones(classes.shape[0], dtype=np.float64, order='C')
+    elif class_weight == 'auto':
+        # proportional to the number of samples in the class
+        weight = np.array([1.0 / np.sum(y == i) for i in classes],
+                          dtype=np.float64, order='C')
+        weight *= classes.shape[0] / np.sum(weight)
+    else:
+        # user-defined dictionary
+        weight = np.ones(classes.shape[0], dtype=np.float64, order='C')
+        if not isinstance(class_weight, dict):
+            raise ValueError("class_weight must be dict, 'auto', or None,"
+                             " got: %r" % class_weight)
+        for c in class_weight:
+            i = np.searchsorted(classes, c)
+            if classes[i] != c:
+                raise ValueError("Class label %d not present." % c)
+            else:
+                weight[i] = class_weight[c]
+
+    return weight