[MRG] Add ChangedBehaviorWarning message for LMNN too (#214)

* Add ChangedBehaviorWarning message for LMNN too

* Remove useless 'as an init'

Author: wdevazelhes

metric_learn/lmnn.py

@@ -17,6 +17,7 @@
 import warnings
 from collections import Counter
 from six.moves import xrange
+from sklearn.exceptions import ChangedBehaviorWarning
 from sklearn.metrics import euclidean_distances
 from sklearn.base import TransformerMixin
 
@@ -26,18 +27,20 @@
 
 # commonality between LMNN implementations
 class _base_LMNN(MahalanobisMixin, TransformerMixin):
-  def __init__(self, init='auto', k=3, min_iter=50, max_iter=1000,
+  def __init__(self, init=None, k=3, min_iter=50, max_iter=1000,
                learn_rate=1e-7, regularization=0.5, convergence_tol=0.001,
                use_pca=True, verbose=False, preprocessor=None,
                n_components=None, num_dims='deprecated', random_state=None):
     """Initialize the LMNN object.
 
     Parameters
     ----------
-    init : string or numpy array, optional (default='auto')
+    init : None, string or numpy array, optional (default=None)
         Initialization of the linear transformation. Possible options are
-        'auto', 'pca', 'lda', 'identity', 'random', and a numpy array of shape
-        (n_features_a, n_features_b).
+        'auto', 'pca', 'identity', 'random', and a numpy array of shape
+        (n_features_a, n_features_b). If None, will be set automatically to
+        'auto' (this option is to raise a warning if 'init' is not set,
+        and stays to its default value None, in v0.5.0).
 
         'auto'
             Depending on ``n_components``, the most reasonable initialization
@@ -135,7 +138,21 @@ def fit(self, X, y):
     if len(label_inds) != num_pts:
       raise ValueError('Must have one label per point.')
     self.labels_ = np.arange(len(unique_labels))
-    self.transformer_ = _initialize_transformer(output_dim, X, y, self.init,
+
+    # if the init is the default (None), we raise a warning
+    if self.init is None:
+      # TODO: replace init=None by init='auto' in v0.6.0 and remove the warning
+      msg = ("Warning, no init was set (`init=None`). As of version 0.5.0, "
+             "the default init will now be set to 'auto', instead of the "
+             "previous identity matrix. If you still want to use the identity "
+             "matrix as before, set init='identity'. This warning "
+             "will disappear in v0.6.0, and `init` parameter's default value "
+             "will be set to 'auto'.")
+      warnings.warn(msg, ChangedBehaviorWarning)
+      init = 'auto'
+    else:
+      init = self.init
+    self.transformer_ = _initialize_transformer(output_dim, X, y, init,
                                                 self.verbose,
                                                 self.random_state)
     required_k = np.bincount(label_inds).min()

metric_learn/lsml.py

@@ -94,7 +94,7 @@ def _fit(self, quadruplets, weights=None):
     else:
       self.w_ = weights
     self.w_ /= self.w_.sum()  # weights must sum to 1
-    # if the prior is the default (identity), we raise a warning just in case
+    # if the prior is the default (None), we raise a warning
     if self.prior is None:
       msg = ("Warning, no prior was set (`prior=None`). As of version 0.5.0, "
              "the default prior will now be set to "

metric_learn/mlkr.py

@@ -156,7 +156,7 @@ def fit(self, X, y):
       m = self.n_components
       if m is None:
           m = d
-      # if the init is the default (identity), we raise a warning just in case
+      # if the init is the default (None), we raise a warning
       if self.init is None:
         # TODO:
         #  replace init=None by init='auto' in v0.6.0 and remove the warning

metric_learn/nca.py

@@ -141,16 +141,16 @@ def fit(self, X, y):
     train_time = time.time()
 
     # Initialize A
-    # if the init is the default (auto), we raise a warning just in case
+    # if the init is the default (None), we raise a warning
     if self.init is None:
       # TODO: replace init=None by init='auto' in v0.6.0 and remove the warning
       msg = ("Warning, no init was set (`init=None`). As of version 0.5.0, "
              "the default init will now be set to 'auto', instead of the "
-             "previous scaling matrix. same scaling matrix as before as an "
-             "init, set init=np.eye(X.shape[1])/"
-             "(np.maximum(X.max(axis=0)-X.min(axis=0), EPS))). This warning "
-             "will disappear in v0.6.0, and `init` parameter's default value "
-             "will be set to 'auto'.")
+             "previous scaling matrix. If you still want to use the same "
+             "scaling matrix as before, set "
+             "init=np.eye(X.shape[1])/(np.maximum(X.max(axis=0)-X.min(axis=0)"
+             ", EPS))). This warning will disappear in v0.6.0, and `init` "
+             "parameter's default value will be set to 'auto'.")
       warnings.warn(msg, ChangedBehaviorWarning)
       init = 'auto'
     else:

metric_learn/sdml.py

@@ -111,7 +111,7 @@ def _fit(self, pairs, y):
                                     type_of_inputs='tuples')
 
     # set up (the inverse of) the prior M
-    # if the prior is the default (identity), we raise a warning just in case
+    # if the prior is the default (None), we raise a warning
     if self.prior is None:
       # TODO:
       #  replace prior=None by prior='identity' in v0.6.0 and remove the

test/metric_learn_test.py

@@ -275,6 +275,23 @@ def grad(x):
                 np.linalg.norm(approx_fprime(L.ravel(), fun, epsilon)))
     np.testing.assert_almost_equal(rel_diff, 0., decimal=5)
 
+  def test_changed_behaviour_warning(self):
+    # test that a ChangedBehavior warning is thrown about the init, if the
+    # default parameters are used.
+    # TODO: remove in v.0.6
+    X = np.array([[0, 0], [0, 1], [2, 0], [2, 1]])
+    y = np.array([1, 0, 1, 0])
+    lmnn = LMNN(k=2)
+    msg = ("Warning, no init was set (`init=None`). As of version 0.5.0, "
+           "the default init will now be set to 'auto', instead of the "
+           "previous identity matrix. If you still want to use the identity "
+           "matrix as before, set init='identity'. This warning "
+           "will disappear in v0.6.0, and `init` parameter's default value "
+           "will be set to 'auto'.")
+    with pytest.warns(ChangedBehaviorWarning) as raised_warning:
+      lmnn.fit(X, y)
+    assert any(msg == str(wrn.message) for wrn in raised_warning)
+
 
 @pytest.mark.parametrize('X, y, loss', [(np.array([[0], [1], [2], [3]]),
                                          [1, 1, 0, 0], 3.0),
@@ -744,11 +761,11 @@ def test_changed_behaviour_warning(self):
     nca = NCA()
     msg = ("Warning, no init was set (`init=None`). As of version 0.5.0, "
            "the default init will now be set to 'auto', instead of the "
-           "previous scaling matrix. same scaling matrix as before as an "
-           "init, set init=np.eye(X.shape[1])/"
-           "(np.maximum(X.max(axis=0)-X.min(axis=0), EPS))). This warning will"
-           " disappear in v0.6.0, and `init` parameter's default value will "
-           "be set to 'auto'.")
+           "previous scaling matrix. If you still want to use the same "
+           "scaling matrix as before, set "
+           "init=np.eye(X.shape[1])/(np.maximum(X.max(axis=0)-X.min(axis=0)"
+           ", EPS))). This warning will disappear in v0.6.0, and `init` "
+           "parameter's default value will be set to 'auto'.")
     with pytest.warns(ChangedBehaviorWarning) as raised_warning:
       nca.fit(X, y)
     assert any(msg == str(wrn.message) for wrn in raised_warning)

test/test_base_metric.py

@@ -21,7 +21,7 @@ def test_covariance(self):
   def test_lmnn(self):
     self.assertRegexpMatches(
       str(metric_learn.LMNN()),
-        r"(python_)?LMNN\(convergence_tol=0.001, init='auto', k=3, "
+        r"(python_)?LMNN\(convergence_tol=0.001, init=None, k=3, "
         r"learn_rate=1e-07,\s+"
         r"max_iter=1000, min_iter=50, n_components=None, "
         r"num_dims='deprecated',\s+preprocessor=None, random_state=None, "