Try deflaking test timing (#32351)

* Try deflaking test timing

* Add another potential timeout

* Simpler technique

sdks/python/apache_beam/ml/inference/base_test.py

@@ -887,13 +887,12 @@ def test_run_inference_timeout_not_hit(self):
   def test_run_inference_timeout_does_garbage_collection(self):
     with tempfile.TemporaryDirectory() as tmp_dirname:
       tmp_path = os.path.join(tmp_dirname, 'tmp_filename')
-      expected_file_contents = 'Deleted FakeSlowModel'
       with TestPipeline() as pipeline:
         # Start with bad example which gets timed out.
         # Then provide plenty of time for GC to happen.
-        examples = [20] + [1] * 15
+        examples = [20] + [1] * 15 + [20, 20, 20]
         expected_good = [1] * 15
-        expected_bad = [20]
+        expected_bad = [20, 20, 20, 20]
         pcoll = pipeline | 'start' >> beam.Create(examples)
         main, other = pcoll | base.RunInference(
             FakeSlowModelHandler(
@@ -910,7 +909,7 @@ def test_run_inference_timeout_does_garbage_collection(self):
 
       with open(tmp_path) as f:
         s = f.read()
-        self.assertEqual(s, expected_file_contents)
+        self.assertNotEqual(s, '')
 
   def test_run_inference_impl_inference_args(self):
     with TestPipeline() as pipeline: