legacy optimizer 5

Author: antoinedemathelin

adapt/parameter_based/_transfer_tree.py

@@ -456,7 +456,11 @@ def _force_coherence(self,rule,node=0,Translate=False,indexes_nodes=list(),drift
             if not coh:
                 if Translate :
                     if auto_drift:
-                        b_infs,b_sups = ut.bounds_rule(rule,self.estimator_.n_features_)
+                        try:
+                            n_feat = self.estimator_.n_features_
+                        except:
+                            n_feat = self.estimator_.n_features_in_
+                        b_infs,b_sups = ut.bounds_rule(rule, n_feat)
                         if non_coherent_sense == -1:
                             if b_sups[phi] == np.inf:
                                 self.updateSplit(node,phi,th+D_MARGIN)
@@ -817,8 +821,12 @@ def _strut(self,X_target_node,Y_target_node,node=0,no_prune_on_cl=False,cl_no_pr
         maj_class = np.argmax(self.estimator_.tree_.value[node, :].copy())
 
         if min_drift is None or max_drift is None:
-            min_drift = np.zeros(self.estimator_.n_features_)
-            max_drift = np.zeros(self.estimator_.n_features_)
+            try:
+                n_feat = self.estimator_.n_features_
+            except:
+                n_feat = self.estimator_.n_features_in_
+            min_drift = np.zeros(self.estimator_.n_feat)
+            max_drift = np.zeros(self.estimator_.n_feat)
 
         current_class_distribution = ut.compute_class_distribution(classes_, Y_target_node)
         is_reached = (Y_target_node.size > 0)

tests/test_finetuning.py

@@ -41,23 +41,23 @@ def test_finetune():
 
     fine_tuned = FineTuning(encoder=model.encoder_, task=model.task_,
                             training=True,
-                            loss="bce", optimizer="adam", random_state=0)
+                            loss="bce", optimizer=Adam(), random_state=0)
     fine_tuned.fit(Xt[ind], yt[ind], epochs=100, verbose=0)
 
     assert np.abs(fine_tuned.encoder_.get_weights()[0] - model.encoder_.get_weights()[0]).sum() > 1.
     assert np.mean((fine_tuned.predict(Xt).ravel()>0.5) == yt) > 0.9
 
     fine_tuned = FineTuning(encoder=model.encoder_, task=model.task_,
                             training=[True, False],
-                            loss="bce", optimizer="adam", random_state=0)
+                            loss="bce", optimizer=Adam(), random_state=0)
     fine_tuned.fit(Xt[ind], yt[ind], epochs=100, verbose=0)
 
     assert np.abs(fine_tuned.encoder_.get_weights()[0] - model.encoder_.get_weights()[0]).sum() == 0.
     assert np.abs(fine_tuned.encoder_.get_weights()[-1] - model.encoder_.get_weights()[-1]).sum() > 1.
 
     fine_tuned = FineTuning(encoder=model.encoder_, task=model.task_,
                             training=[False],
-                            loss="bce", optimizer="adam", random_state=0)
+                            loss="bce", optimizer=Adam(), random_state=0)
     fine_tuned.fit(Xt[ind], yt[ind], epochs=100, verbose=0)
 
     assert np.abs(fine_tuned.encoder_.get_weights()[0] - model.encoder_.get_weights()[0]).sum() == 0.
@@ -66,13 +66,13 @@ def test_finetune():
 
 def test_finetune_pretrain():
     model = FineTuning(encoder=encoder, task=task, pretrain=True, pretrain__epochs=2,
-                       loss="bce", optimizer="adam", random_state=0)
+                       loss="bce", optimizer=Adam(), random_state=0)
     model.fit(Xs, ys, epochs=1, verbose=0)
 
 
 def test_clone():
     model = FineTuning(encoder=encoder, task=task,
-                       loss="bce", optimizer="adam", random_state=0)
+                       loss="bce", optimizer=Adam(), random_state=0)
     model.fit(Xs, ys, epochs=1, verbose=0)
 
     new_model = clone(model)

tests/test_iwc.py

@@ -7,6 +7,10 @@
 from adapt.utils import make_classification_da
 from adapt.instance_based import IWC
 from adapt.utils import get_default_discriminator
+try:
+    from tensorflow.keras.optimizers.legacy import Adam
+except:
+    from tensorflow.keras.optimizers import Adam
 
 Xs, ys, Xt, yt = make_classification_da()
 
@@ -34,7 +38,7 @@ def test_default_classif():
 
 def test_nn_classif():
     model = IWC(RidgeClassifier(0.), classifier=get_default_discriminator(),
-                cl_params=dict(epochs=10, optimizer="adam", loss="bce", verbose=0),
+                cl_params=dict(epochs=10, optimizer=Adam(), loss="bce", verbose=0),
                 Xt=Xt, random_state=0)
     model.fit(Xs, ys);
     model.predict(Xt)

tests/test_iwn.py

@@ -7,6 +7,10 @@
 from adapt.instance_based import IWN
 from adapt.utils import get_default_task
 from sklearn.neighbors import KNeighborsClassifier
+try:
+    from tensorflow.keras.optimizers.legacy import Adam
+except:
+    from tensorflow.keras.optimizers import Adam
 
 Xs, ys, Xt, yt = make_classification_da()
 
@@ -21,7 +25,7 @@ def test_iwn():
 
 def test_iwn_fit_estim():
     task = get_default_task()
-    task.compile(optimizer="adam", loss="mse", metrics=["mae"])
+    task.compile(optimizer=Adam(), loss="mse", metrics=["mae"])
     model = IWN(task, Xt=Xt, sigma_init=0.1, random_state=0,
                 pretrain=True, pretrain__epochs=100, pretrain__verbose=0)
     model.fit(Xs, ys)