Make Average Model support for 'moving mean' and 'moving variance' of batch_normal op

python/paddle/fluid/optimizer.py

@@ -850,22 +850,38 @@ class ModelAverage(Optimizer):
         self.min_average_window = min_average_window
         self.max_average_window = max_average_window
         self.params_grads = params_grads
+
+        # append 'moving mean' and 'moving variance' to self.params_grads
+        pattern = re.compile(r"batch_norm_\d+\.w_[1,2]")
+        for param in framework.default_main_program().global_block(
+        ).all_parameters():
+            if pattern.match(param.name) is not None:
+                self.params_grads.append((param, None))
+        # create a tmp gradient variable to backup parameter value
+        # for parameter whose grad is None
+        for i, param_grad in enumerate(self.params_grads):
+            param, grad = param_grad
+            if grad is None:
+                grad = param.block.create_var(
+                    name=unique_name.generate(".".join([param.name, 'tmp'])),
+                    dtype=param.dtype,
+                    persistable=False,
+                    stop_gradient=stop_gradient)
+                self.params_grads[i] = (param, grad)
+
         for param, grad in self.params_grads:
-            if grad is not None:
             self._append_average_accumulate_op(param)
 
         self.apply_program = Program()
         block = self.apply_program.global_block()
         with program_guard(main_program=self.apply_program):
             for param_grad in self.params_grads:
-                if param_grad[1] is not None:
                 self._add_average_apply_op(block, param_grad)
 
         self.restore_program = Program()
         block = self.restore_program.global_block()
         with program_guard(main_program=self.restore_program):
             for param_grad in self.params_grads:
-                if param_grad[1] is not None:
                 self._add_average_restore_op(block, param_grad)
 
     def _add_average_apply_op(self, block, param_grad):