Add additional E2E tests for examples

docs/source/tutorials/create-a-new-workflow.md

@@ -286,7 +286,7 @@ uv pip install -e examples/documentation_guides/workflows/text_file_ingest
 Run the workflow with the following command:
 ```bash
 nat run --config_file examples/documentation_guides/workflows/text_file_ingest/configs/config.yml \
-   --input "What does DOCA GPUNetIO to remove the CPU from the critical path?"
+   --input "What does DOCA GPUNetIO do to remove the CPU from the critical path?"
 ```
 
 If successful, you should receive output similar to the following:

examples/advanced_agents/profiler_agent/tests/test_profiler_agent.py

@@ -94,4 +94,6 @@ async def test_token_usage_tool(df_path: Path):
 @pytest.mark.usefixtures("nvidia_api_key")
 async def test_full_workflow():
     config_file: Path = locate_example_config(ProfilerAgentConfig)
-    await run_workflow(config_file, "Is the product of 33 * 4 greater than the current hour of the day?", "yes")
+    await run_workflow(config_file=config_file,
+                       question="Is the product of 33 * 4 greater than the current hour of the day?",
+                       expected_answer="yes")

examples/agents/tests/test_agents.py

@@ -62,7 +62,7 @@ def rewoo_answer_fixture(request: pytest.FixtureRequest, rewoo_data: list[dict])
                          indirect=True)
 async def test_rewoo_full_workflow(rewoo_question: str, rewoo_answer: str):
     config_file = os.path.join(AGENTS_DIR, "rewoo/configs/config.yml")
-    await run_workflow(config_file, rewoo_question, rewoo_answer)
+    await run_workflow(config_file=config_file, question=rewoo_question, expected_answer=rewoo_answer)
 
 
 @pytest.mark.slow
@@ -79,4 +79,4 @@ async def test_rewoo_full_workflow(rewoo_question: str, rewoo_answer: str):
     ],
     ids=["mixture_of_agents", "react", "react-reasoning", "tool_calling", "tool_calling-reasoning"])
 async def test_agent_full_workflow(config_file: str, question: str, answer: str):
-    await run_workflow(config_file, question, answer)
+    await run_workflow(config_file=config_file, question=question, expected_answer=answer)

examples/custom_functions/automated_description_generation/tests/test_auto_desc_generation.py

@@ -37,4 +37,4 @@ async def test_full_workflow(milvus_uri: str) -> None:
     config.retrievers['retriever'].uri = HttpUrl(url=milvus_uri)
 
     # Unfortunately the workflow itself returns inconsistent results
-    await run_workflow(None, "List 5 subspecies of Aardvark?", "Aardvark", config=config)
+    await run_workflow(config=config, question="List 5 subspecies of Aardvark?", expected_answer="Aardvark")

examples/documentation_guides/tests/test_custom_workflow.py

@@ -44,7 +44,7 @@ def answer_fixture() -> str:
 @pytest.mark.usefixtures("nvidia_api_key")
 async def test_custom_full_workflow(custom_workflow_dir: Path, question: str, answer: str):
     config_file = custom_workflow_dir / "custom_config.yml"
-    await run_workflow(config_file, question, answer)
+    await run_workflow(config_file=config_file, question=question, expected_answer=answer)
 
 
 @pytest.mark.slow
@@ -53,4 +53,4 @@ async def test_custom_full_workflow(custom_workflow_dir: Path, question: str, an
 async def test_search_full_workflow(custom_workflow_dir: Path, question: str, answer: str):
     # Technically this is the same as the custom workflow test, but it requires a second key
     config_file = custom_workflow_dir / "search_config.yml"
-    await run_workflow(config_file, question, answer)
+    await run_workflow(config_file=config_file, question=question, expected_answer=answer)

examples/documentation_guides/tests/test_text_file_ingest.py

@@ -0,0 +1,67 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import logging
+import sys
+from collections.abc import Generator
+from pathlib import Path
+
+import pytest
+
+from nat.test.utils import locate_example_config
+from nat.test.utils import run_workflow
+
+logger = logging.getLogger(__name__)
+
+
+@pytest.fixture(name="text_file_ingest_dir", scope="session")
+def text_file_ingest_dir_fixture(workflows_dir: Path) -> Path:
+    text_file_ingest = workflows_dir / "text_file_ingest"
+    assert text_file_ingest.exists(), f"Could not find text_file_ingest example at {text_file_ingest}"
+    return text_file_ingest
+
+
+@pytest.fixture(name="src_dir", scope="session", autouse=True)
+def src_dir_fixture(text_file_ingest_dir: Path) -> Path:
+    src_dir = text_file_ingest_dir / "src"
+    assert src_dir.exists(), f"Could not find text_file_ingest src at {src_dir}"
+
+    return src_dir
+
+
+@pytest.fixture(name="add_src_dir_to_path", scope="session")
+def add_src_dir_to_path_fixture(src_dir: Path) -> Generator[str]:
+    # Since this is a documentation guide, it is not installed by default, so we need to manually append it to the path
+    abs_src_dir = str(src_dir.absolute())
+    if abs_src_dir not in sys.path:
+        added = True
+        sys.path.append(abs_src_dir)
+    else:
+        added = False
+
+    yield abs_src_dir
+
+    if added:
+        sys.path.remove(abs_src_dir)
+
+
+@pytest.mark.integration
+@pytest.mark.usefixtures("nvidia_api_key", "add_src_dir_to_path")
+async def test_text_file_ingest_full_workflow():
+    from text_file_ingest.text_file_ingest_function import TextFileIngestFunctionConfig
+    config_file = locate_example_config(TextFileIngestFunctionConfig)
+    await run_workflow(config_file=config_file,
+                       question="What does DOCA GPUNetIO do to remove the CPU from the critical path?",
+                       expected_answer="GPUDirect")

examples/documentation_guides/workflows/text_file_ingest/src/text_file_ingest/configs/config.yml

@@ -29,19 +29,6 @@ llms:
     _type: nim
     model_name: meta/llama-3.1-70b-instruct
     temperature: 0.0
-  nim_rag_eval_llm:
-    _type: nim
-    model_name: meta/llama-3.1-70b-instruct
-    max_tokens: 8
-  nim_rag_eval_large_llm:
-    _type: nim
-    model_name: meta/llama-3.1-70b-instruct
-    max_tokens: 2048
-  nim_trajectory_eval_llm:
-    _type: nim
-    model_name: meta/llama-3.1-70b-instruct
-    temperature: 0.0
-    max_tokens: 1024
 
 embedders:
   nv-embedqa-e5-v5:
@@ -54,36 +41,3 @@ workflow:
   llm_name: nim_llm
   verbose: true
   parse_agent_response_max_retries: 3
-
-eval:
-  general:
-    output_dir: .tmp/nat/examples/getting_started/simple_web_query/
-    dataset:
-      _type: json
-      file_path: examples/evaluation_and_profiling/simple_web_query_eval/data/langsmith.json
-    profiler:
-      fit_model: True
-
-  evaluators:
-    rag_accuracy:
-      _type: ragas
-      metric: AnswerAccuracy
-      llm_name: nim_rag_eval_llm
-    rag_groundedness:
-      _type: ragas
-      metric: ResponseGroundedness
-      llm_name: nim_rag_eval_llm
-    rag_relevance:
-      _type: ragas
-      metric: ContextRelevance
-      llm_name: nim_rag_eval_llm
-    rag_factual_correctness:
-      _type: ragas
-      metric:
-        FactualCorrectness:
-          kwargs:
-            mode: precision
-      llm_name: nim_rag_eval_large_llm # requires more tokens
-    trajectory:
-      _type: trajectory
-      llm_name: nim_trajectory_eval_llm

examples/evaluation_and_profiling/email_phishing_analyzer/configs

@@ -0,0 +1 @@
+src/nat_email_phishing_analyzer/configs

examples/evaluation_and_profiling/email_phishing_analyzer/data

@@ -0,0 +1 @@
+src/nat_email_phishing_analyzer/data

examples/evaluation_and_profiling/email_phishing_analyzer/tests/test_email_phishing_analyzer.py

@@ -0,0 +1,63 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import logging
+from pathlib import Path
+
+import pytest
+
+from nat.test.utils import locate_example_config
+from nat.test.utils import run_workflow
+
+logger = logging.getLogger(__name__)
+
+
+@pytest.mark.integration
+@pytest.mark.usefixtures("nvidia_api_key")
+async def test_run_full_workflow():
+    from nat.runtime.loader import load_config
+    from nat_email_phishing_analyzer.register import EmailPhishingAnalyzerConfig
+
+    config_file: Path = locate_example_config(EmailPhishingAnalyzerConfig)
+    config = load_config(config_file)
+
+    # Unfortunately the workflow itself returns inconsistent results
+    await run_workflow(
+        config=config,
+        question=(
+            "Dear [Customer], Thank you for your purchase on [Date]. We have processed a refund of $[Amount] to your "
+            "account. Please provide your account and routing numbers so we can complete the transaction. Thank you, "
+            "[Your Company]"),
+        expected_answer="likely")
+
+
+@pytest.mark.skip(reason="This test gets rate limited potentially issue #842 and does not complete")
+@pytest.mark.integration
+@pytest.mark.usefixtures("nvidia_api_key", "require_nest_asyncio")
+async def test_optimize_full_workflow(capsys):
+    from nat.data_models.config import Config
+    from nat.data_models.optimizer import OptimizerRunConfig
+    from nat.profiler.parameter_optimization.optimizer_runtime import optimize_config
+    from nat_email_phishing_analyzer.register import EmailPhishingAnalyzerConfig
+
+    config_file: Path = locate_example_config(EmailPhishingAnalyzerConfig, "config_optimizer.yml")
+    config = OptimizerRunConfig(config_file=config_file,
+                                dataset=None,
+                                override=(('eval.general.max_concurrency', '1'), ('optimizer.numeric.n_trials', '1')))
+    optimized_config = await optimize_config(config)
+    assert isinstance(optimized_config, Config)
+    captured_output = capsys.readouterr()
+
+    assert "All optimization phases complete" in captured_output.out

examples/evaluation_and_profiling/simple_calculator_eval/README.md

@@ -47,6 +47,8 @@ Install this evaluation example:
 uv pip install -e examples/evaluation_and_profiling/simple_calculator_eval
 ```
 
+> **Note**: If you encounter rate limiting (`[429] Too Many Requests`) during evaluation, try setting the `eval.general.max_concurrency` value either in the YAML directly or via the command line with: `--override eval.general.max_concurrency 1`.
+
 ## Run the Workflow
 
 ### Running Evaluation

examples/evaluation_and_profiling/simple_calculator_eval/src/nat_simple_calculator_eval/configs/config-tunable-rag-eval.yml

@@ -72,7 +72,7 @@ workflow:
 
 eval:
   general:
-    output_dir: .tmp/nat/examples/getting_started/simple_web_query
+    output_dir: .tmp/nat/examples/getting_started/simple_calculator
     dataset:
       _type: json
       file_path: examples/getting_started/simple_calculator/data/simple_calculator.json

examples/evaluation_and_profiling/simple_calculator_eval/tests/test_simple_calculator_eval.py

@@ -0,0 +1,79 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import logging
+from pathlib import Path
+
+import pytest
+
+from nat.eval.evaluate import EvaluationRun
+from nat.eval.evaluate import EvaluationRunConfig
+from nat.test.utils import locate_example_config
+from nat.test.utils import validate_workflow_output
+
+logger = logging.getLogger(__name__)
+
+
+@pytest.mark.integration
+@pytest.mark.usefixtures("nvidia_api_key")
+async def test_eval():
+    """
+    1. nat-eval writes the workflow output to workflow_output.json
+    2. nat-eval creates a file with scores for each evaluation metric.
+    3. This test audits -
+       a. the rag accuracy metric
+       b. the trajectory score (if present)
+    """
+    import nat_simple_calculator_eval
+
+    # Get config dynamically
+    config_file: Path = locate_example_config(nat_simple_calculator_eval, "config-tunable-rag-eval.yml")
+
+    # Create the configuration object for running the evaluation, single rep using the eval config in eval_config.yml
+    # WIP: skip test if eval config is not present
+    config = EvaluationRunConfig(
+        config_file=config_file,
+        dataset=None,
+        result_json_path="$",
+        skip_workflow=False,
+        skip_completed_entries=False,
+        endpoint=None,
+        endpoint_timeout=30,
+        reps=1,
+        override=(('eval.general.max_concurrency', '1'), ),
+    )
+
+    # Run evaluation
+    eval_runner = EvaluationRun(config=config)
+    output = await eval_runner.run_and_evaluate()
+
+    # Ensure the workflow was not interrupted
+    assert not output.workflow_interrupted, "The workflow was interrupted"
+
+    # Look for the tuneable_eval_output file
+    tuneable_eval_output: Path | None = None
+
+    for output_file in output.evaluator_output_files:
+        assert output_file.exists()
+        output_file_str = str(output_file)
+        if "tuneable_eval_output" in output_file_str:
+            tuneable_eval_output = output_file
+
+    # Validate the workflow output
+    assert output.workflow_output_file, "The workflow_output.json file was not created"
+    validate_workflow_output(output.workflow_output_file)
+
+    # Verify that atleast one tuneable_eval_output file is present
+    assert tuneable_eval_output, "Expected output file does not exist"

examples/evaluation_and_profiling/simple_web_query_eval/tests/test_simple_web_query_eval.py

@@ -19,39 +19,14 @@
 
 import pytest
 
-import nat_simple_web_query_eval
 from nat.eval.evaluate import EvaluationRun
 from nat.eval.evaluate import EvaluationRunConfig
 from nat.test.utils import locate_example_config
+from nat.test.utils import validate_workflow_output
 
 logger = logging.getLogger(__name__)
 
 
-def validate_workflow_output(workflow_output_file: Path):
-    """
-    Validate the contents of the workflow output file.
-    WIP: output format should be published as a schema and this validation should be done against that schema.
-    """
-    # Ensure the workflow_output.json file was created
-    assert workflow_output_file.exists(), "The workflow_output.json file was not created"
-
-    # Read and validate the workflow_output.json file
-    try:
-        with open(workflow_output_file, encoding="utf-8") as f:
-            result_json = json.load(f)
-    except json.JSONDecodeError:
-        pytest.fail("Failed to parse workflow_output.json as valid JSON")
-
-    assert isinstance(result_json, list), "The workflow_output.json file is not a list"
-    assert len(result_json) > 0, "The workflow_output.json file is empty"
-    assert isinstance(result_json[0], dict), "The workflow_output.json file is not a list of dictionaries"
-
-    # Ensure required keys exist
-    required_keys = ["id", "question", "answer", "generated_answer", "intermediate_steps"]
-    for key in required_keys:
-        assert all(item.get(key) for item in result_json), f"The '{key}' key is missing in workflow_output.json"
-
-
 def validate_rag_accuracy(rag_metric_output_file: Path, score: float):
     """
     1. Validate the contents of the rag evaluator ouput file.
@@ -110,6 +85,8 @@ async def test_eval():
        a. the rag accuracy metric
        b. the trajectory score (if present)
     """
+    import nat_simple_web_query_eval
+
     # Get config dynamically
     config_file: Path = locate_example_config(nat_simple_web_query_eval, "eval_config.yml")