DEP: Add deprecation warnings for num_labels

Author: William de Vazelhes

metric_learn/itml.py

@@ -14,6 +14,7 @@
 """
 
 from __future__ import print_function, absolute_import
+import warnings
 import numpy as np
 from six.moves import xrange
 from sklearn.metrics import pairwise_distances
@@ -143,7 +144,8 @@ def metric(self):
 class ITML_Supervised(ITML):
   """Information Theoretic Metric Learning (ITML)"""
   def __init__(self, gamma=1., max_iter=1000, convergence_threshold=1e-3,
-               num_constraints=None, bounds=None, A0=None, verbose=False):
+               num_labeled='deprecated', num_constraints=None, bounds=None,
+               A0=None, verbose=False):
     """Initialize the supervised version of `ITML`.
 
     `ITML_Supervised` creates pairs of similar sample by taking same class
@@ -156,6 +158,10 @@ def __init__(self, gamma=1., max_iter=1000, convergence_threshold=1e-3,
         value for slack variables
     max_iter : int, optional
     convergence_threshold : float, optional
+    num_labeled : Not used
+          .. deprecated:: 0.4.0
+        `num_labeled` was deprecated in version 0.4.0 and will
+        be removed in 0.5.0.
     num_constraints: int, optional
         number of constraints to generate
     bounds : list (pos,neg) pairs, optional
@@ -164,10 +170,12 @@ def __init__(self, gamma=1., max_iter=1000, convergence_threshold=1e-3,
         initial regularization matrix, defaults to identity
     verbose : bool, optional
         if True, prints information while learning
+        learning_rate : Not used
     """
     ITML.__init__(self, gamma=gamma, max_iter=max_iter,
                   convergence_threshold=convergence_threshold,
                   A0=A0, verbose=verbose)
+    self.num_labeled = num_labeled
     self.num_constraints = num_constraints
     self.bounds = bounds
 
@@ -185,6 +193,10 @@ def fit(self, X, y, random_state=np.random):
     random_state : numpy.random.RandomState, optional
         If provided, controls random number generation.
     """
+    if self.num_labeled != 'deprecated':
+      warnings.warn('"num_labeled" parameter is not used.'
+                    ' It has been deprecated in version 0.4 and will be'
+                    'removed in 0.5', DeprecationWarning)
     X, y = check_X_y(X, y)
     num_constraints = self.num_constraints
     if num_constraints is None:

metric_learn/lsml.py

@@ -8,6 +8,7 @@
 """
 
 from __future__ import print_function, absolute_import, division
+import warnings
 import numpy as np
 import scipy.linalg
 from six.moves import xrange
@@ -133,7 +134,8 @@ def _gradient(self, metric):
 
 class LSML_Supervised(LSML):
   def __init__(self, tol=1e-3, max_iter=1000, prior=None,
-               num_constraints=None, weights=None, verbose=False):
+               num_labeled='deprecated', num_constraints=None, weights=None,
+               verbose=False):
     """Initialize the supervised version of `LSML`.
 
     `LSML_Supervised` creates quadruplets from labeled samples by taking two
@@ -147,6 +149,10 @@ def __init__(self, tol=1e-3, max_iter=1000, prior=None,
     max_iter : int, optional
     prior : (d x d) matrix, optional
         guess at a metric [default: covariance(X)]
+    num_labeled : Not used
+      .. deprecated:: 0.4.0
+         `num_labeled` was deprecated in version 0.4.0 and will
+         be removed in 0.5.0.
     num_constraints: int, optional
         number of constraints to generate
     weights : (m,) array of floats, optional
@@ -156,6 +162,7 @@ def __init__(self, tol=1e-3, max_iter=1000, prior=None,
     """
     LSML.__init__(self, tol=tol, max_iter=max_iter, prior=prior,
                   verbose=verbose)
+    self.num_labeled = num_labeled
     self.num_constraints = num_constraints
     self.weights = weights
 
@@ -173,6 +180,10 @@ def fit(self, X, y, random_state=np.random):
     random_state : numpy.random.RandomState, optional
         If provided, controls random number generation.
     """
+    if self.num_labeled != 'deprecated':
+      warnings.warn('"num_labeled" parameter is not used.'
+                    ' It has been deprecated in version 0.4 and will be'
+                    'removed in 0.5', DeprecationWarning)
     X, y = check_X_y(X, y)
     num_constraints = self.num_constraints
     if num_constraints is None:

metric_learn/mmc.py

@@ -17,6 +17,7 @@
 """
 
 from __future__ import print_function, absolute_import, division
+import warnings
 import numpy as np
 from six.moves import xrange
 from sklearn.metrics import pairwise_distances
@@ -384,8 +385,8 @@ def transformer(self):
 class MMC_Supervised(MMC):
   """Mahalanobis Metric for Clustering (MMC)"""
   def __init__(self, max_iter=100, max_proj=10000, convergence_threshold=1e-6,
-               num_constraints=None, A0=None, diagonal=False,
-               diagonal_c=1.0, verbose=False):
+               num_labeled='deprecated', num_constraints=None, A0=None,
+               diagonal=False, diagonal_c=1.0, verbose=False):
     """Initialize the supervised version of `MMC`.
 
     `MMC_Supervised` creates pairs of similar sample by taking same class
@@ -397,6 +398,10 @@ def __init__(self, max_iter=100, max_proj=10000, convergence_threshold=1e-6,
     max_iter : int, optional
     max_proj : int, optional
     convergence_threshold : float, optional
+    num_labeled : Not used
+      .. deprecated:: 0.4.0
+         `num_labeled` was deprecated in version 0.4.0 and will
+         be removed in 0.5.0.
     num_constraints: int, optional
         number of constraints to generate
     A0 : (d x d) matrix, optional
@@ -415,6 +420,7 @@ def __init__(self, max_iter=100, max_proj=10000, convergence_threshold=1e-6,
                  convergence_threshold=convergence_threshold,
                  A0=A0, diagonal=diagonal, diagonal_c=diagonal_c,
                  verbose=verbose)
+    self.num_labeled = num_labeled
     self.num_constraints = num_constraints
 
   def fit(self, X, y, random_state=np.random):
@@ -429,6 +435,10 @@ def fit(self, X, y, random_state=np.random):
     random_state : numpy.random.RandomState, optional
         If provided, controls random number generation.
     """
+    if self.num_labeled != 'deprecated':
+      warnings.warn('"num_labeled" parameter is not used.'
+                    ' It has been deprecated in version 0.4 and will be'
+                    'removed in 0.5', DeprecationWarning)
     X, y = check_X_y(X, y)
     num_constraints = self.num_constraints
     if num_constraints is None:

metric_learn/sdml.py

@@ -9,6 +9,7 @@
 """
 
 from __future__ import absolute_import
+import warnings
 import numpy as np
 from scipy.sparse.csgraph import laplacian
 from sklearn.covariance import graph_lasso
@@ -82,7 +83,7 @@ def fit(self, X, W):
 
 class SDML_Supervised(SDML):
   def __init__(self, balance_param=0.5, sparsity_param=0.01, use_cov=True,
-               num_constraints=None, verbose=False):
+               num_labeled='deprecated', num_constraints=None, verbose=False):
     """Initialize the supervised version of `SDML`.
 
     `SDML_Supervised` creates pairs of similar sample by taking same class
@@ -97,6 +98,10 @@ def __init__(self, balance_param=0.5, sparsity_param=0.01, use_cov=True,
         trade off between optimizer and sparseness (see graph_lasso)
     use_cov : bool, optional
         controls prior matrix, will use the identity if use_cov=False
+    num_labeled : Not used
+      .. deprecated:: 0.4.0
+         `num_labeled` was deprecated in version 0.4.0 and will
+         be removed in 0.5.0.
     num_constraints : int, optional
         number of constraints to generate
     verbose : bool, optional
@@ -105,6 +110,7 @@ def __init__(self, balance_param=0.5, sparsity_param=0.01, use_cov=True,
     SDML.__init__(self, balance_param=balance_param,
                   sparsity_param=sparsity_param, use_cov=use_cov,
                   verbose=verbose)
+    self.num_labeled = num_labeled
     self.num_constraints = num_constraints
 
   def fit(self, X, y, random_state=np.random):
@@ -125,6 +131,10 @@ def fit(self, X, y, random_state=np.random):
     self : object
         Returns the instance.
     """
+    if self.num_labeled != 'deprecated':
+      warnings.warn('"num_labeled" parameter is not used.'
+                    ' It has been deprecated in version 0.4 and will be'
+                    'removed in 0.5', DeprecationWarning)
     y = check_array(y, ensure_2d=False)
     num_constraints = self.num_constraints
     if num_constraints is None:

test/metric_learn_test.py

@@ -55,6 +55,17 @@ def test_iris(self):
     csep = class_separation(lsml.transform(), self.iris_labels)
     self.assertLess(csep, 0.8)  # it's pretty terrible
 
+  def test_deprecation(self):
+    # test that the right deprecation message is thrown.
+    # TODO: remove in v.0.5
+    X = np.array([[0, 0], [0, 1], [2, 0], [2, 1]])
+    y = np.array([1, 0, 1, 0])
+    lsml_supervised = LSML_Supervised(num_labeled=np.inf)
+    msg = ('"num_labeled" parameter is not used.'
+           ' It has been deprecated in version 0.4 and will be'
+           'removed in 0.5')
+    assert_warns_message(DeprecationWarning, msg, lsml_supervised.fit, X, y)
+
 
 class TestITML(MetricTestCase):
   def test_iris(self):
@@ -64,6 +75,17 @@ def test_iris(self):
     csep = class_separation(itml.transform(), self.iris_labels)
     self.assertLess(csep, 0.2)
 
+  def test_deprecation(self):
+    # test that the right deprecation message is thrown.
+    # TODO: remove in v.0.5
+    X = np.array([[0, 0], [0, 1], [2, 0], [2, 1]])
+    y = np.array([1, 0, 1, 0])
+    itml_supervised = ITML_Supervised(num_labeled=np.inf)
+    msg = ('"num_labeled" parameter is not used.'
+           ' It has been deprecated in version 0.4 and will be'
+           'removed in 0.5')
+    assert_warns_message(DeprecationWarning, msg, itml_supervised.fit, X, y)
+
 
 class TestLMNN(MetricTestCase):
   def test_iris(self):
@@ -118,6 +140,17 @@ def test_iris(self):
     csep = class_separation(sdml.transform(), self.iris_labels)
     self.assertLess(csep, 0.25)
 
+  def test_deprecation(self):
+    # test that the right deprecation message is thrown.
+    # TODO: remove in v.0.5
+    X = np.array([[0, 0], [0, 1], [2, 0], [2, 1]])
+    y = np.array([1, 0, 1, 0])
+    sdml_supervised = SDML_Supervised(num_labeled=np.inf)
+    msg = ('"num_labeled" parameter is not used.'
+           ' It has been deprecated in version 0.4 and will be'
+           'removed in 0.5')
+    assert_warns_message(DeprecationWarning, msg, sdml_supervised.fit, X, y)
+
 
 class TestNCA(MetricTestCase):
   def test_iris(self):
@@ -343,6 +376,17 @@ def test_iris(self):
     csep = class_separation(mmc.transform(), self.iris_labels)
     self.assertLess(csep, 0.2)
 
+  def test_deprecation(self):
+    # test that the right deprecation message is thrown.
+    # TODO: remove in v.0.5
+    X = np.array([[0, 0], [0, 1], [2, 0], [2, 1]])
+    y = np.array([1, 0, 1, 0])
+    mmc_supervised = MMC_Supervised(num_labeled=np.inf)
+    msg = ('"num_labeled" parameter is not used.'
+           ' It has been deprecated in version 0.4 and will be'
+           'removed in 0.5')
+    assert_warns_message(DeprecationWarning, msg, mmc_supervised.fit, X, y)
+
 
 @pytest.mark.parametrize(('algo_class', 'dataset'),
                          [(NCA, make_classification()),

test/test_base_metric.py

@@ -30,25 +30,27 @@ def test_itml(self):
 """.strip('\n'))
     self.assertEqual(str(metric_learn.ITML_Supervised()), """
 ITML_Supervised(A0=None, bounds=None, convergence_threshold=0.001, gamma=1.0,
-        max_iter=1000, num_constraints=None, verbose=False)
+        max_iter=1000, num_constraints=None, num_labeled='deprecated',
+        verbose=False)
 """.strip('\n'))
 
   def test_lsml(self):
     self.assertEqual(
         str(metric_learn.LSML()),
         "LSML(max_iter=1000, prior=None, tol=0.001, verbose=False)")
     self.assertEqual(str(metric_learn.LSML_Supervised()), """
-LSML_Supervised(max_iter=1000, num_constraints=None, prior=None, tol=0.001,
-        verbose=False, weights=None)
+LSML_Supervised(max_iter=1000, num_constraints=None, num_labeled='deprecated',
+        prior=None, tol=0.001, verbose=False, weights=None)
 """.strip('\n'))
 
   def test_sdml(self):
     self.assertEqual(str(metric_learn.SDML()),
                      "SDML(balance_param=0.5, sparsity_param=0.01, "
                      "use_cov=True, verbose=False)")
     self.assertEqual(str(metric_learn.SDML_Supervised()), """
-SDML_Supervised(balance_param=0.5, num_constraints=None, sparsity_param=0.01,
-        use_cov=True, verbose=False)
+SDML_Supervised(balance_param=0.5, num_constraints=None,
+        num_labeled='deprecated', sparsity_param=0.01, use_cov=True,
+        verbose=False)
 """.strip('\n'))
 
   def test_rca(self):
@@ -71,7 +73,7 @@ def test_mmc(self):
     self.assertEqual(str(metric_learn.MMC_Supervised()), """
 MMC_Supervised(A0=None, convergence_threshold=1e-06, diagonal=False,
         diagonal_c=1.0, max_iter=100, max_proj=10000, num_constraints=None,
-        verbose=False)
+        num_labeled='deprecated', verbose=False)
 """.strip('\n'))
 
 if __name__ == '__main__':