fix: wrap evaluate command with if __name__ == "__main__" (#644)

Signed-off-by: Marta Stepniewska-Dziubinska <martas@nvidia.com>

Author: marta-sd

.github/config/requirements.txt

@@ -1,4 +1,4 @@
-sphinx
+sphinx==8.2.3
 sphinx-autobuild    # For live doc serving while editing docs
 sphinx-autodoc2     # For documenting Python API
 autodoc_pydantic    # For rendering pydantic dataclasses

docs/get-started/_snippets/core_basic.py

@@ -19,8 +19,6 @@
 # Prerequisites: Set your API key
 # export NGC_API_KEY="nvapi-..."
 
-import os
-
 # [snippet-start]
 from nemo_evaluator.api.api_dataclasses import (
     ApiEndpoint,
@@ -49,14 +47,9 @@
     )
 )
 
-# Run evaluation
-result = evaluate(eval_cfg=eval_config, target_cfg=target_config)
-print(f"Evaluation completed: {result}")
-# [snippet-end]
 
 if __name__ == "__main__":
-    # Note: This requires a valid API key to actually run
-    api_key = os.getenv("NGC_API_KEY")
-    if not api_key:
-        print("Set NGC_API_KEY environment variable to run this example")
-        print("export NGC_API_KEY='your-key-here'")
+    # Run evaluation
+    result = evaluate(eval_cfg=eval_config, target_cfg=target_config)
+    print(f"Evaluation completed: {result}")
+# [snippet-end]

docs/get-started/_snippets/core_full_example.py

@@ -52,15 +52,9 @@
     )
 )
 
-# Run evaluation
-try:
+if __name__ == "__main__":
+    # Run evaluation
     result = evaluate(eval_cfg=eval_config, target_cfg=target_config)
     print(f"Evaluation completed. Results saved to: {eval_config.output_dir}")
-except Exception as e:
-    print(f"Evaluation failed: {e}")
-# [snippet-end]
 
-if __name__ == "__main__":
-    print(
-        "Replace 'nvapi-your-key-here' with your actual NGC API key to run this example"
-    )
+# [snippet-end]

docs/get-started/_snippets/core_multi_benchmark.py

@@ -39,21 +39,19 @@
     )
 )
 
-# Run multiple benchmarks
-benchmarks = ["mmlu_pro", "humaneval", "mgsm"]
-results = {}
 
-for benchmark in benchmarks:
-    config = EvaluationConfig(
-        type=benchmark,
-        output_dir=f"./results/{benchmark}",
-        params=ConfigParams(limit_samples=10),
-    )
+if __name__ == "__main__":
+    # Run multiple benchmarks
+    benchmarks = ["mmlu_pro", "humaneval", "mgsm"]
+    results = {}
 
-    result = evaluate(eval_cfg=config, target_cfg=target_config)
-    results[benchmark] = result
-# [snippet-end]
+    for benchmark in benchmarks:
+        config = EvaluationConfig(
+            type=benchmark,
+            output_dir=f"./results/{benchmark}",
+            params=ConfigParams(limit_samples=10),
+        )
 
-if __name__ == "__main__":
-    print("Multi-benchmark evaluation example")
-    print("Replace 'your_api_key_here' with your actual API key to run")
+        result = evaluate(eval_cfg=config, target_cfg=target_config)
+        results[benchmark] = result
+# [snippet-end]

docs/get-started/_snippets/nemo_fw_basic.py

@@ -41,7 +41,8 @@
     ),
 )
 
-# Run evaluation
-results = evaluate(target_cfg=target, eval_cfg=config)
-print(results)
+if __name__ == "__main__":
+    # Run evaluation
+    results = evaluate(target_cfg=target, eval_cfg=config)
+    print(results)
 # [snippet-end]

docs/get-started/quickstart/core.md

@@ -166,8 +166,9 @@ config = EvaluationConfig(
     )
 )
 
-result = evaluate(eval_cfg=config, target_cfg=target)
-print(f"Evaluation completed: {result}")
+if __name__ == "__main__":
+    result = evaluate(eval_cfg=config, target_cfg=target)
+    print(f"Evaluation completed: {result}")
 ```
 
 **Available Interceptors:**