Moving sgd to optimizers

apex/optimizers/__init__.py

 from .fused_adam import FusedAdam
+from .fused_sgd import FusedSGD
 from .fp16_optimizer import FP16_Optimizer

apex/optim/sgd.py → apex/optimizers/fused_sgd.py

@@ -3,7 +3,7 @@ from torch.optim.optimizer import Optimizer, required
 
 from apex.multi_tensor_apply import multi_tensor_applier
 
-class SGD(Optimizer):
+class FusedSGD(Optimizer):
     r"""Implements stochastic gradient descent (optionally with momentum).
 
     Nesterov momentum is based on the formula from