!32938 optimize global step for optimizer

Merge pull request !32938 from zhangbuxue/optimize_global_step_for_optimizer

mindspore/python/mindspore/nn/optim/asgd.py

@@ -177,6 +177,8 @@ class ASGD(Optimizer):
         gradients = self.gradients_centralization(gradients)
         gradients = self.scale_grad(gradients)
         lrs = self.get_lr()
+        if not self.is_dynamic_lr_or_weight_decay():
+            self.assignadd(self.global_step, self.global_step_increase_tensor)
         success = True
 
         for index, (grad, param, mu, eta, ax) in enumerate(zip(gradients, self.parameters, self.mu, self.eta, self.ax)):

mindspore/python/mindspore/nn/optim/lamb.py

@@ -355,6 +355,8 @@ class Lamb(Optimizer):
     def construct(self, gradients):
         weight_decay = self.get_weight_decay()
         lr = self.get_lr()
+        if not self.is_dynamic_lr_or_weight_decay():
+            self.assignadd(self.global_step, self.global_step_increase_tensor)
         lamb_opt = _lamb_opt_ascend if self.device_ascend else _lamb_opt
         gradients = self.gradients_centralization(gradients)
         if self.is_group:

mindspore/python/mindspore/nn/optim/lars.py

@@ -187,4 +187,6 @@ class LARS(Optimizer):
             gradients = self.hyper_map(F.partial(_lars_opt, self.lars, self.loss_scale, lr, weight_decay),
                                        gradients, params, self.decay_flags, self.lars_flag)
         success = self.opt(gradients)
+        if self.is_dynamic_lr_or_weight_decay() and not self.opt.is_dynamic_lr_or_weight_decay():
+            self.assignadd(self.global_step, self.global_step_increase_tensor)
         return success

mindspore/python/mindspore/nn/optim/optimizer.py

@@ -675,7 +675,8 @@ class Optimizer(Cell):
                     lr += (current_dynamic_lr,)
             else:
                 lr = self.learning_rate(self.global_step)
-        self.assignadd(self.global_step, self.global_step_increase_tensor)
+        if self.is_dynamic_lr_or_weight_decay():
+            self.assignadd(self.global_step, self.global_step_increase_tensor)
         return lr
 
     def get_lr_parameter(self, param):
@@ -732,6 +733,15 @@ class Optimizer(Cell):
 
         return lr if isinstance(param, list) else lr[0]
 
+    def is_dynamic_lr_or_weight_decay(self):
+        """
+        Determine whether the learning rate or weight decay is dynamic.
+
+        Returns:
+             bool, represents the learning rate or weight decay is dynamic or not.
+        """
+        return self.dynamic_lr or self.dynamic_weight_decay
+
     def _get_parameter_group_id(self):
         """
         Get the parameter partition group id, which is less than the number of devices.

mindspore/python/mindspore/nn/optim/rprop.py

@@ -192,6 +192,8 @@ class Rprop(Optimizer):
         gradients = self.gradients_centralization(gradients)
         gradients = self.scale_grad(gradients)
         lrs = self.get_lr()
+        if not self.is_dynamic_lr_or_weight_decay():
+            self.assignadd(self.global_step, self.global_step_increase_tensor)
         success = True
 
         for index, (grad, param, prev, step_size) in enumerate(zip(gradients, self.parameters,

tests/st/dump/test_cell_dump.py

@@ -146,7 +146,7 @@ def test_ascend_not_cell_dump():
         check_dump_structure(dump_path, dump_config_path, 1, 1, 1)
 
         # make sure set_dump is ignored and all cell layer are dumped
-        assert len(os.listdir(dump_file_path)) == 11
+        assert len(os.listdir(dump_file_path)) == 10
         del os.environ['MINDSPORE_DUMP_CONFIG']