Fix windows runtime warning problem

Fix windows runtime warning problem

Fix windows runtime warning issue

Author: Shaobo-Zhou

src/mqt/predictor/ml/predictor.py

@@ -191,7 +191,10 @@ def _compile_all_circuits_devicewise(
             if (path_compiled_circuits / (target_filename + ".qasm")).exists():
                 continue
             try:
-                res = timeout_watcher(rl_compile, [qc, device, self.figure_of_merit, rl_pred], timeout)
+                if sys.platform == "win32":
+                    res = rl_compile(qc, device, self.figure_of_merit, rl_pred)
+                else:
+                    res = timeout_watcher(rl_compile, [qc, device, self.figure_of_merit, rl_pred], timeout)
                 if isinstance(res, tuple):
                     compiled_qc = res[0]
                     with Path(path_compiled_circuits / (target_filename + ".qasm")).open("w", encoding="utf-8") as f:

tests/hellinger_distance/test_estimated_hellinger_distance.py

@@ -11,7 +11,6 @@
 from __future__ import annotations
 
 import re
-import sys
 import warnings
 from pathlib import Path
 from typing import TYPE_CHECKING
@@ -217,15 +216,9 @@ def test_train_and_qcompile_with_hellinger_model(source_path: Path, target_path:
                 dump(qc, f)
 
         # Generate compiled circuits (using trained RL model)
-        if sys.platform == "win32":
-            with pytest.warns(RuntimeWarning, match=re.escape("Timeout is not supported on Windows.")):
-                ml_predictor.compile_training_circuits(
-                    timeout=6000, path_compiled_circuits=target_path, path_uncompiled_circuits=source_path
-                )
-        else:
-            ml_predictor.compile_training_circuits(
-                timeout=6000, path_compiled_circuits=target_path, path_uncompiled_circuits=source_path
-            )
+        ml_predictor.compile_training_circuits(
+            timeout=6000, path_compiled_circuits=target_path, path_uncompiled_circuits=source_path
+        )
 
         # Generate training data from the compiled circuits
         ml_predictor.generate_training_data(path_uncompiled_circuits=source_path, path_compiled_circuits=target_path)