fix: Fix: Correctly define choices as tuple for reward-model arg Fixes #941 (#974)

Flink-ddd · web-flow · commit 86aeab2bba5a · 2025-06-08T22:42:57.000-07:00
Signed-off-by: Vensenmu &lt;vensenmu@gmail.com&gt;
diff --git a/applications/DeepSpeed-Chat/e2e_rlhf.py b/applications/DeepSpeed-Chat/e2e_rlhf.py
@@ -65,7 +65,7 @@ def parse_args():
         "--reward-model",
         type=lambda x: x.replace("facebook/opt-", ""),
         default="350m",
-        choices=("350m"),
+        choices=("350m", ),
         help="Which facebook/opt-* model to use for Reward (step 2)",
     )
     parser.add_argument(

Original file line number	Diff line number	Diff line change
`@@ -65,7 +65,7 @@ def parse_args():`
`65`	`65`	`"--reward-model",`
`66`	`66`	`type=lambda x: x.replace("facebook/opt-", ""),`
`67`	`67`	`default="350m",`
`68`		`- choices=("350m"),`
	`68`	`+ choices=("350m", ),`
`69`	`69`	`help="Which facebook/opt-* model to use for Reward (step 2)",`
`70`	`70`	`)`
`71`	`71`	`parser.add_argument(`