bf16_optimizer: fixes to different grad acc dtype (#6485)

- fix step function to cast to FP32 before step in case of different
gradient accumulation data type
- remove redundatn function initialize_optimizer_states()

deepspeed/runtime/bf16_optimizer.py

@@ -197,10 +197,6 @@ class BF16_Optimizer(ZeROOptimizer):
 
             see_memory_usage(f'after initializing group {i}', force=True)
 
-        see_memory_usage('before initialize_optimizer', force=True)
-        self.initialize_optimizer_states()
-        see_memory_usage('end initialize_optimizer', force=True)
-
         self._grad_acc_hooks = []
         if self.immediate_grad_update:
             self.create_grad_acc_hooks()
@@ -252,25 +248,6 @@ class BF16_Optimizer(ZeROOptimizer):
                                                     self.optimizer.state)
             self._hp_optimizer_states_linked = True
 
-    def initialize_optimizer_states(self):
-        """Take an optimizer step with zero-valued gradients to allocate internal
-        optimizer state.
-
-        This helps prevent memory fragmentation by allocating optimizer state at the
-        beginning of training instead of after activations have been allocated.
-        """
-        for param_partition, grad_partition in zip(self.fp32_groups_flat_partition,
-                                                   self.fp32_groups_gradient_flat_partition):
-            # In case of grad acc dtype different than FP32, need to cast to high precision.
-            param_partition.grad = grad_partition.to(
-                param_partition.dtype) if grad_partition.dtype != param_partition.dtype else grad_partition
-
-        if self.grad_acc_dtype is not torch.float32:
-            for param_partition in self.fp32_groups_flat_partition:
-                param_partition.grad = None
-
-        self.clear_hp_grads()
-
     def _split_flat_tensor(self, flat_tensor, num_elem_list):
         assert sum(num_elem_list) <= flat_tensor.numel()
         tensor_list = []
@@ -317,8 +294,18 @@ class BF16_Optimizer(ZeROOptimizer):
                                         mpu=self.mpu,
                                         use_graph=self.graph_harvesting)
 
+        for param_partition, grad_partition in zip(self.fp32_groups_flat_partition,
+                                                   self.fp32_groups_gradient_flat_partition):
+            # In case of grad acc dtype different than FP32, need to cast to high precision.
+            param_partition.grad = grad_partition.to(
+                param_partition.dtype) if grad_partition.dtype != param_partition.dtype else grad_partition
+
         self.optimizer.step()
 
+        if self.grad_acc_dtype is not torch.float32:
+            for param_partition in self.fp32_groups_flat_partition:
+                param_partition.grad = None
+
         # We need to link optimizer state after the first step() call
         self._lazy_init_hp_params_optimizer_state()