Skip to content

Commit cf21ec8

Browse files
authored
Update deployment.py
1 parent b9618d8 commit cf21ec8

File tree

1 file changed

+8
-2
lines changed

1 file changed

+8
-2
lines changed

ads/aqua/modeldeployment/deployment.py

Lines changed: 8 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -921,9 +921,15 @@ def _create(
921921
{"MODEL": f"{AQUA_MODEL_DEPLOYMENT_FOLDER}{aqua_model.base_model_id}/"}
922922
)
923923

924+
base_model_inference_key = aqua_model.base_model_id
925+
for item in aqua_model.member_models:
926+
if item["model_id"] == aqua_model.base_model_id:
927+
base_model_inference_key = item["inference_key"]
928+
break
929+
924930
params_dict = get_params_dict(params)
925-
# updates `--served-model-name` with service model id
926-
params_dict.update({"--served-model-name": aqua_model.base_model_id})
931+
# updates `--served-model-name` with service model inference key
932+
params_dict.update({"--served-model-name": base_model_inference_key})
927933
# TODO: sets `--max-lora-rank` as 32 in params for now, will revisit later
928934
params_dict.update({"--max-lora-rank": 32})
929935
# adds `--enable_lora` to parameters

0 commit comments

Comments
 (0)