Remove decay variable from optimizers.

PiperOrigin-RevId: 346172198
Change-Id: I2f98d52d47ffe81ee3baa5891e187c2c84f145cf

tensorflow/python/keras/optimizer_v2/adadelta.py

@@ -153,7 +153,7 @@ class Adadelta(optimizer_v2.OptimizerV2):
     config = super(Adadelta, self).get_config()
     config.update({
         'learning_rate': self._serialize_hyperparameter('learning_rate'),
-        'decay': self._serialize_hyperparameter('decay'),
+        'decay': self._initial_decay,
         'rho': self._serialize_hyperparameter('rho'),
         'epsilon': self.epsilon,
     })

tensorflow/python/keras/optimizer_v2/adagrad.py

@@ -157,7 +157,7 @@ class Adagrad(optimizer_v2.OptimizerV2):
     config = super(Adagrad, self).get_config()
     config.update({
         'learning_rate': self._serialize_hyperparameter('learning_rate'),
-        'decay': self._serialize_hyperparameter('decay'),
+        'decay': self._initial_decay,
         'initial_accumulator_value': self._initial_accumulator_value,
         'epsilon': self.epsilon,
     })

tensorflow/python/keras/optimizer_v2/adam.py

@@ -244,7 +244,7 @@ class Adam(optimizer_v2.OptimizerV2):
     config = super(Adam, self).get_config()
     config.update({
         'learning_rate': self._serialize_hyperparameter('learning_rate'),
-        'decay': self._serialize_hyperparameter('decay'),
+        'decay': self._initial_decay,
         'beta_1': self._serialize_hyperparameter('beta_1'),
         'beta_2': self._serialize_hyperparameter('beta_2'),
         'epsilon': self.epsilon,
@@ -468,7 +468,7 @@ class NonFusedAdam(optimizer_v2.OptimizerV2):
     config = super(NonFusedAdam, self).get_config()
     config.update({
         'learning_rate': self._serialize_hyperparameter('learning_rate'),
-        'decay': self._serialize_hyperparameter('decay'),
+        'decay': self._initial_decay,
         'beta_1': self._serialize_hyperparameter('beta_1'),
         'beta_2': self._serialize_hyperparameter('beta_2'),
         'epsilon': self.epsilon,

tensorflow/python/keras/optimizer_v2/adamax.py

@@ -180,7 +180,7 @@ class Adamax(optimizer_v2.OptimizerV2):
     config = super(Adamax, self).get_config()
     config.update({
         'learning_rate': self._serialize_hyperparameter('learning_rate'),
-        'decay': self._serialize_hyperparameter('decay'),
+        'decay': self._initial_decay,
         'beta_1': self._serialize_hyperparameter('beta_1'),
         'beta_2': self._serialize_hyperparameter('beta_2'),
         'epsilon': self.epsilon,

tensorflow/python/keras/optimizer_v2/ftrl.py

@@ -234,7 +234,7 @@ class Ftrl(optimizer_v2.OptimizerV2):
         'learning_rate':
             self._serialize_hyperparameter('learning_rate'),
         'decay':
-            self._serialize_hyperparameter('decay'),
+            self._initial_decay,
         'initial_accumulator_value':
             self._initial_accumulator_value,
         'learning_rate_power':

tensorflow/python/keras/optimizer_v2/gradient_descent.py

@@ -187,7 +187,7 @@ class SGD(optimizer_v2.OptimizerV2):
     config = super(SGD, self).get_config()
     config.update({
         "learning_rate": self._serialize_hyperparameter("learning_rate"),
-        "decay": self._serialize_hyperparameter("decay"),
+        "decay": self._initial_decay,
         "momentum": self._serialize_hyperparameter("momentum"),
         "nesterov": self.nesterov,
     })

tensorflow/python/keras/optimizer_v2/nadam.py

@@ -214,7 +214,7 @@ class Nadam(optimizer_v2.OptimizerV2):
     config = super(Nadam, self).get_config()
     config.update({
         'learning_rate': self._serialize_hyperparameter('learning_rate'),
-        'decay': self._serialize_hyperparameter('decay'),
+        'decay': self._initial_decay,
         'beta_1': self._serialize_hyperparameter('beta_1'),
         'beta_2': self._serialize_hyperparameter('beta_2'),
         'epsilon': self.epsilon,

tensorflow/python/keras/optimizer_v2/optimizer_v2.py

@@ -1004,7 +1004,7 @@ class OptimizerV2(trackable.Trackable):
       lr_t = math_ops.cast(lr_t(local_step), var_dtype)
     if self._initial_decay > 0.:
       local_step = math_ops.cast(self.iterations, var_dtype)
-      decay_t = self._get_hyper("decay", var_dtype)
+      decay_t = math_ops.cast(self._initial_decay, var_dtype)
       lr_t = lr_t / (1. + decay_t * local_step)
     return lr_t
 

tensorflow/python/keras/optimizer_v2/optimizer_v2_test.py

@@ -908,6 +908,19 @@ class OptimizerWithFunctionTest(test.TestCase, parameterized.TestCase):
     self.assertAllClose([0., 1.], fn(), atol=1e-4)
     self.assertAllClose([-1, 0.], fn(), atol=1e-4)
 
+  def testBasicWithConstantDecay(self):
+    var = variables.Variable([1.0, 2.0], dtype=dtypes.float32)
+    loss = lambda: 3 * var
+    opt = adam.Adam(learning_rate=1.0)
+
+    @def_function.function
+    def fn():
+      opt.minimize(loss, [var])
+      return var
+
+    self.assertAllClose([0., 1.], fn(), atol=1e-4)
+    self.assertAllClose([-1, 0.], fn(), atol=1e-4)
+
   def testVarKeyWithVarCreatedInEager(self):
     a = variables.Variable([1., 2.], name='var')
     b = variables.Variable([1.], name='var')

tensorflow/python/keras/optimizer_v2/rmsprop.py

@@ -290,7 +290,7 @@ class RMSprop(optimizer_v2.OptimizerV2):
     config = super(RMSprop, self).get_config()
     config.update({
         "learning_rate": self._serialize_hyperparameter("learning_rate"),
-        "decay": self._serialize_hyperparameter("decay"),
+        "decay": self._initial_decay,
         "rho": self._serialize_hyperparameter("rho"),
         "momentum": self._serialize_hyperparameter("momentum"),
         "epsilon": self.epsilon,