optimize performance for pynative

mindcv/optim/adamw.py

@@ -9,6 +9,11 @@
 from mindspore.nn.optim import Optimizer
 from mindspore.nn.optim.optimizer import opt_init_args_register
 
+try:
+    from mindspore import jit
+except ImportError:
+    from mindspore import ms_function as jit
+
 
 def _check_param_value(beta1, beta2, eps, prim_name):
     """Check the type of inputs."""
@@ -154,6 +159,7 @@ def __init__(
         self.reciprocal_scale = Tensor(1.0 / loss_scale, ms.float32)
         self.clip = clip
 
+    @jit
     def construct(self, gradients):
         lr = self.get_lr()
         gradients = scale_grad(gradients, self.reciprocal_scale)

mindcv/optim/adan.py

@@ -5,6 +5,11 @@
 from mindspore.common.tensor import Tensor
 from mindspore.nn.optim.optimizer import Optimizer, opt_init_args_register
 
+try:
+    from mindspore import jit
+except ImportError:
+    from mindspore import ms_function as jit
+
 _adan_opt = ops.MultitypeFuncGraph("adan_opt")
 
 
@@ -144,6 +149,7 @@ def __init__(
 
         self.weight_decay = Tensor(weight_decay, mstype.float32)
 
+    @jit
     def construct(self, gradients):
         params = self._parameters
         moment1 = self.moment1

mindcv/optim/lion.py

@@ -8,6 +8,11 @@
 from mindspore.nn.optim import Optimizer
 from mindspore.nn.optim.optimizer import opt_init_args_register
 
+try:
+    from mindspore import jit
+except ImportError:
+    from mindspore import ms_function as jit
+
 
 def _check_param_value(beta1, beta2, prim_name):
     """Check the type of inputs."""
@@ -142,6 +147,7 @@ def __init__(
         self.reciprocal_scale = Tensor(1.0 / loss_scale, ms.float32)
         self.clip = clip
 
+    @jit
     def construct(self, gradients):
         lr = self.get_lr()
         gradients = scale_grad(gradients, self.reciprocal_scale)

mindcv/optim/nadam.py

@@ -9,6 +9,11 @@
 from mindspore.nn.optim import Optimizer
 from mindspore.nn.optim.optimizer import opt_init_args_register
 
+try:
+    from mindspore import jit
+except ImportError:
+    from mindspore import ms_function as jit
+
 
 def _check_param_value(beta1, beta2, eps, prim_name):
     """Check the type of inputs."""
@@ -48,6 +53,7 @@ def __init__(
         self.mu_schedule = Parameter(initializer(1, [1], ms.float32), name="mu_schedule")
         self.beta2_power = Parameter(initializer(1, [1], ms.float32), name="beta2_power")
 
+    @jit
     def construct(self, gradients):
         lr = self.get_lr()
         params = self.parameters