updating tests

Author: nnshah1

src/c++/perf_analyzer/genai-perf/genai_perf/llm_inputs/llm_inputs.py

@@ -225,7 +225,6 @@ def _get_input_dataset_from_url(
         url = cls._resolve_url(dataset_name)
         configured_url = cls._create_configured_url(url, starting_index, length)
         dataset = cls._download_dataset(configured_url)
-
         return dataset
 
     @classmethod

src/c++/perf_analyzer/genai-perf/genai_perf/parser.py

@@ -119,7 +119,7 @@ def _check_conditional_args(
     if args.service_kind != "triton":
         if args.output_tokens_mean_deterministic:
             parser.error(
-                "The --output-tokens-mean-deterministic option is only supported with the Triton service-kind."
+                "The --output-tokens-mean-deterministic option is only supported with the kserve endpoint type."
             )
 
     return args
@@ -252,7 +252,7 @@ def _add_input_args(parser):
         help=f"When using --output-tokens-mean, this flag can be set to "
         "improve precision by setting the minimum number of tokens "
         "equal to the requested number of tokens. This is currently "
-        "supported with the Triton service-kind. "
+        "supported with the kserve endpoint type. "
         "Note that there is still some variability in the requested number "
         "of output tokens, but GenAi-Perf attempts its best effort with your "
         "model to get the right number of output tokens. ",

src/c++/perf_analyzer/genai-perf/tests/test_cli.py

@@ -376,7 +376,7 @@ def test_unrecognized_arg(self, monkeypatch, capsys):
                     "100",
                     "--output-tokens-mean-deterministic",
                 ],
-                "The --output-tokens-mean-deterministic option is only supported with the Triton service-kind",
+                "The --output-tokens-mean-deterministic option is only supported with the kserve endpoint type",
             ),
         ],
     )

src/c++/perf_analyzer/genai-perf/tests/test_wrapper.py

@@ -43,7 +43,7 @@ class TestWrapper:
         ],
     )
     def test_url_exactly_once_triton(self, monkeypatch, arg):
-        args = ["genai-perf", "-m", "test_model", "--service-kind", "triton"] + arg
+        args = ["genai-perf", "-m", "test_model", "--endpoint-type", "kserve"] + arg
         monkeypatch.setattr("sys.argv", args)
         args, extra_args = parser.parse_args()
         cmd = Profiler.build_cmd(args, extra_args)
@@ -70,7 +70,7 @@ def test_url_exactly_once_triton(self, monkeypatch, arg):
         ],
     )
     def test_profile_export_filepath(self, monkeypatch, arg, expected_filepath):
-        args = ["genai-perf", "-m", "test_model", "--service-kind", "triton"] + arg
+        args = ["genai-perf", "-m", "test_model", "--endpoint-type", "kserve"] + arg
         monkeypatch.setattr("sys.argv", args)
         args, extra_args = parser.parse_args()
         cmd = Profiler.build_cmd(args, extra_args)
@@ -87,7 +87,7 @@ def test_profile_export_filepath(self, monkeypatch, arg, expected_filepath):
         ],
     )
     def test_service_triton(self, monkeypatch, arg):
-        args = ["genai-perf", "-m", "test_model", "--service-kind", "triton"] + arg
+        args = ["genai-perf", "-m", "test_model", "--endpoint-type", "kserve"] + arg
         monkeypatch.setattr("sys.argv", args)
         args, extra_args = parser.parse_args()
         cmd = Profiler.build_cmd(args, extra_args)
@@ -113,8 +113,6 @@ def test_service_openai(self, monkeypatch, arg):
             "genai-perf",
             "-m",
             "test_model",
-            "--service-kind",
-            "openai",
         ] + arg
         monkeypatch.setattr("sys.argv", args)
         args, extra_args = parser.parse_args()