remove all ensemble_model references, replace with executor_model

examples/Serving-An-Implicit-Model-With-Merlin-Systems.ipynb

@@ -412,7 +412,7 @@
     "]\n",
     "# send request to tritonserver\n",
     "with grpcclient.InferenceServerClient(\"localhost:8001\") as client:\n",
-    "    response = client.infer(\"ensemble_model\", inputs, outputs=outputs)"
+    "    response = client.infer(\"executor_model\", inputs, outputs=outputs)"
    ]
   },
   {

examples/Serving-An-XGboost-Model-With-Merlin-Systems.ipynb

@@ -455,7 +455,7 @@
     "]\n",
     "# send request to tritonserver\n",
     "with grpcclient.InferenceServerClient(\"localhost:8001\") as client:\n",
-    "    response = client.infer(\"ensemble_model\", inputs, outputs=outputs)"
+    "    response = client.infer(\"executor_model\", inputs, outputs=outputs)"
    ]
   },
   {

examples/Serving-Ranking-Models-With-Merlin-Systems.ipynb

@@ -1074,7 +1074,7 @@
    "source": [
     "# send request to tritonserver\n",
     "with grpcclient.InferenceServerClient(\"localhost:8001\") as client:\n",
-    "    response = client.infer(\"ensemble_model\", inputs, request_id=\"1\", outputs=outputs)"
+    "    response = client.infer(\"executor_model\", inputs, request_id=\"1\", outputs=outputs)"
    ]
   },
   {

merlin/systems/triton/utils.py

@@ -169,7 +169,7 @@ def send_triton_request(
     request_id : str, optional
         The id of the inference request, by default "1"
     triton_model : str, optional
-        Name of the model to run inputs through, by default "ensemble_model"
+        Name of the model to run inputs through, by default "executor_model"
 
     Returns
     -------

tests/integration/t4r/test_pytorch_backend.py

@@ -117,7 +117,7 @@ def test_serve_t4r_with_torchscript(tmpdir):
     # Send request to Triton and check response
     # ===========================================
     triton_response = run_ensemble_on_tritonserver(
-        tmpdir, input_schema, df, output_schema.column_names, "ensemble_model"
+        tmpdir, input_schema, df, output_schema.column_names, "executor_model"
     )
 
     assert triton_response

tests/unit/examples/test_serving_ranking_models_with_merlin_systems.py

@@ -112,6 +112,6 @@ def test_example_04_exporting_ranking_models(tb):
     )
 
     response = run_ensemble_on_tritonserver(
-        "/tmp/data/ensemble/", schema.without(["click"]), batch, outputs, "ensemble_model"
+        "/tmp/data/ensemble/", schema.without(["click"]), batch, outputs, "executor_model"
     )
     assert len(response["click/binary_classification_task"]) == 3

tests/unit/systems/dag/ops/test_ops.py

@@ -51,7 +51,7 @@ def test_softmax_sampling(tmpdir):
     ens_config, node_configs = ensemble.export(tmpdir)
 
     response = run_ensemble_on_tritonserver(
-        tmpdir, request_schema, request_df, ensemble.output_schema.column_names, "ensemble_model"
+        tmpdir, request_schema, request_df, ensemble.output_schema.column_names, "executor_model"
     )
     assert response is not None
     assert len(response["ordered_ids"]) == 10
@@ -83,7 +83,7 @@ def test_filter_candidates_with_triton(tmpdir):
     ens_config, node_configs = ensemble.export(tmpdir)
 
     response = run_ensemble_on_tritonserver(
-        tmpdir, request_schema, request_df, ensemble.output_schema.column_names, "ensemble_model"
+        tmpdir, request_schema, request_df, ensemble.output_schema.column_names, "executor_model"
     )
 
     assert response is not None

tests/unit/systems/dag/runtimes/triton/test_triton.py

@@ -38,7 +38,6 @@
 @pytest.mark.parametrize(
     ["runtime", "model_name", "expected_model_name"],
     [
-        (None, None, "ensemble_model"),
         (TritonExecutorRuntime(), None, "executor_model"),
         (TritonExecutorRuntime(), "triton_model", "triton_model"),
     ],

tests/unit/systems/ops/tf/test_ensemble.py

@@ -96,7 +96,7 @@ def test_workflow_tf_e2e_config_verification(tmpdir, dataset, engine):
         parsed = text_format.Parse(raw_config, config)
 
         # The config file contents are correct
-        assert parsed.name == "ensemble_model"
+        assert parsed.name == "executor_model"
         assert parsed.platform == "ensemble"
         assert hasattr(parsed, "ensemble_scheduling")
 
@@ -145,7 +145,7 @@ def test_workflow_tf_e2e_multi_op_run(tmpdir, dataset, engine):
 
     # Creating Triton Ensemble Config
     ensemble_config, nodes_config = triton_ens.export(str(tmpdir))
-    config_path = tmpdir / "ensemble_model" / "config.pbtxt"
+    config_path = tmpdir / "executor_model" / "config.pbtxt"
 
     # Checking Triton Ensemble Config
     with open(config_path, "rb") as f:
@@ -154,7 +154,7 @@ def test_workflow_tf_e2e_multi_op_run(tmpdir, dataset, engine):
         parsed = text_format.Parse(raw_config, config)
 
         # The config file contents are correct
-        assert parsed.name == "ensemble_model"
+        assert parsed.name == "executor_model"
         assert parsed.platform == "ensemble"
         assert hasattr(parsed, "ensemble_scheduling")
 
@@ -202,7 +202,7 @@ def test_workflow_tf_python_wrapper(tmpdir, dataset, engine, python):
 
     # Creating Triton Ensemble Config
     ensemble_config, nodes_config = triton_ens.export(str(tmpdir))
-    config_path = tmpdir / "ensemble_model" / "config.pbtxt"
+    config_path = tmpdir / "executor_model" / "config.pbtxt"
 
     # Checking Triton Ensemble Config
     with open(config_path, "rb") as f:
@@ -211,7 +211,7 @@ def test_workflow_tf_python_wrapper(tmpdir, dataset, engine, python):
         parsed = text_format.Parse(raw_config, config)
 
         # The config file contents are correct
-        assert parsed.name == "ensemble_model"
+        assert parsed.name == "executor_model"
         assert parsed.platform == "ensemble"
         assert hasattr(parsed, "ensemble_scheduling")