add selfplay test

add selfplay test

Author: huangshiyu13

.github/workflows/unit_test.yml

@@ -31,7 +31,7 @@ jobs:
       - name: do_unittest
         timeout-minutes: 40
         run: |
-          xvfb-run -s "-screen 0 1400x900x24" python3 -m pytest tests --cov=openrl --cov-report=xml -m unittest --cov-report=term-missing --durations=0 -v --color=yes
+          xvfb-run -s "-screen 0 1400x900x24" python3 -m pytest tests --cov=openrl --cov-report=xml -m unittest --cov-report=term-missing --durations=0 -v --color=yes -s
       - name: Upload coverage reports to Codecov with GitHub Action
         uses: codecov/codecov-action@v3
         with:

examples/selfplay/selfplay.yaml

@@ -1,6 +1,6 @@
 globals:
   selfplay_api_host: 127.0.0.1
-  selfplay_api_port: 10086
+  selfplay_api_port: 13486
 
 seed: 0
 selfplay_api:

openrl/selfplay/callbacks/selfplay_api.py

@@ -50,14 +50,17 @@ def _init_callback(self) -> None:
         )
 
         self.bind = SelfplayAPIServer.bind()
-        serve.run(self.bind)
+        serve.run(self.bind, route_prefix="/selfplay")
         success = False
         try_time = 10
         while not success:
             success = self.api_client.set_sample_strategy(self.sample_strategy)
             try_time -= 1
             if try_time <= 0:
-                raise RuntimeError("Failed to set sample strategy.")
+                raise RuntimeError(
+                    f"Failed to set sample strategy: {self.sample_strategy}. host:"
+                    f" {self.host}, port: {self.port}"
+                )
 
     def _on_step(self) -> bool:
         # print("To send request to API server.")
@@ -72,5 +75,6 @@ def _on_training_end(self) -> None:
             print(f"deleting {application_name}")
         serve.delete(application_name)
         del self.bind
+        serve.shutdown()
         if self.verbose >= 2:
             print(f"delete {application_name} done!")

openrl/selfplay/opponents/utils.py

@@ -28,6 +28,9 @@
 
 
 def check_opponent_template(opponent_template: Union[str, Path]):
+    assert isinstance(opponent_template, Path) or isinstance(
+        opponent_template, str
+    ), f"opponent_template {opponent_template} must be a Path or str"
     if isinstance(opponent_template, str):
         opponent_template = Path(opponent_template)
     assert (

openrl/selfplay/selfplay_api/selfplay_api.py

@@ -33,7 +33,7 @@
 from openrl.selfplay.selfplay_api.opponent_model import BattleResult
 
 
-@serve.deployment(route_prefix="/selfplay")
+@serve.deployment()
 @serve.ingress(app)
 class SelfplayAPIServer(BaseSelfplayAPIServer):
     @app.post("/set_sample_strategy")