Remove extra references to adamw_hf

Author: Rocketknight1

src/transformers/trainer.py

@@ -1420,7 +1420,7 @@ def optimizer_hook(param):
         if args.optim == OptimizerNames.ADAFACTOR:
             optimizer_cls = Adafactor
             optimizer_kwargs.update({"scale_parameter": False, "relative_step": False})
-        elif args.optim in [OptimizerNames.ADAMW_TORCH, OptimizerNames.ADAMW_TORCH_FUSED, OptimizerNames.ADAMW_HF]:
+        elif args.optim in [OptimizerNames.ADAMW_TORCH, OptimizerNames.ADAMW_TORCH_FUSED]:
             from torch.optim import AdamW
 
             optimizer_cls = AdamW

src/transformers/training_args.py

@@ -145,7 +145,6 @@ class OptimizerNames(ExplicitEnum):
     Stores the acceptable string identifiers for optimizers.
     """
 
-    ADAMW_HF = "adamw_hf"
     ADAMW_TORCH = "adamw_torch"
     ADAMW_TORCH_FUSED = "adamw_torch_fused"
     ADAMW_TORCH_XLA = "adamw_torch_xla"
@@ -627,7 +626,7 @@ class TrainingArguments:
 
             The options should be separated by whitespaces.
         optim (`str` or [`training_args.OptimizerNames`], *optional*, defaults to `"adamw_torch"`):
-            The optimizer to use, such as "adamw_hf", "adamw_torch", "adamw_torch_fused", "adamw_apex_fused", "adamw_anyprecision",
+            The optimizer to use, such as "adamw_torch", "adamw_torch_fused", "adamw_apex_fused", "adamw_anyprecision",
             "adafactor". See `OptimizerNames` in [training_args.py](https://github.com/huggingface/transformers/blob/main/src/transformers/training_args.py)
             for a full list of optimizers.
         optim_args (`str`, *optional*):
@@ -2979,7 +2978,7 @@ def set_optimizer(
 
         Args:
             name (`str` or [`training_args.OptimizerNames`], *optional*, defaults to `"adamw_torch"`):
-                The optimizer to use: `"adamw_hf"`, `"adamw_torch"`, `"adamw_torch_fused"`, `"adamw_apex_fused"`,
+                The optimizer to use: `"adamw_torch"`, `"adamw_torch_fused"`, `"adamw_apex_fused"`,
                 `"adamw_anyprecision"` or `"adafactor"`.
             learning_rate (`float`, *optional*, defaults to 5e-5):
                 The initial learning rate.

tests/trainer/test_trainer.py

@@ -5174,16 +5174,6 @@ def hp_name(trial):
     }
 
     optim_test_params = [
-        (
-            OptimizerNames.ADAMW_HF,
-            torch.optim.AdamW,
-            default_adam_kwargs,
-        ),
-        (
-            OptimizerNames.ADAMW_HF.value,
-            torch.optim.AdamW,
-            default_adam_kwargs,
-        ),
         (
             OptimizerNames.ADAMW_TORCH,
             torch.optim.AdamW,