.

agential/cog/strategies/react/code.py

@@ -148,7 +148,9 @@ def generate_action(
 
         return action_type, query
 
-    def generate_observation(self, idx: int, action_type: str, query: str) -> Tuple[str, Dict[str, Any]]:
+    def generate_observation(
+        self, idx: int, action_type: str, query: str
+    ) -> Tuple[str, Dict[str, Any]]:
         """Generates an observation based on the action type and query.
 
         Args:
@@ -188,7 +190,12 @@ def generate_observation(self, idx: int, action_type: str, query: str) -> Tuple[
         return obs, external_tool_info
 
     def create_output_dict(
-        self, thought: str, action_type: str, query: str, obs: str, external_tool_info: Dict[str, Any]
+        self,
+        thought: str,
+        action_type: str,
+        query: str,
+        obs: str,
+        external_tool_info: Dict[str, Any],
     ) -> Dict[str, Any]:
         """Creates a dictionary of the output components.
 

tests/cog/strategies/react/test_code.py

@@ -115,7 +115,9 @@ def test_generate_observation() -> None:
     query = "def first_repeated_char(s):\n    char_set = set()\n    for char in s:\n        if char in char_set:\n            return char\n        else:\n            char_set.add(char)\n    return None"
     llm = FakeListChatModel(responses=[])
     strategy = ReActCodeStrategy(llm=llm)
-    obs, external_tool_info = strategy.generate_observation(idx=0, action_type=action_type, query=query)
+    obs, external_tool_info = strategy.generate_observation(
+        idx=0, action_type=action_type, query=query
+    )
     assert obs == gt_obs
     assert strategy._answer == query
     assert strategy._finished is False
@@ -130,7 +132,9 @@ def test_generate_observation() -> None:
     llm = FakeListChatModel(responses=[])
     strategy = ReActCodeStrategy(llm=llm)
     strategy._answer = "print('Hello World')"
-    obs, external_tool_info = strategy.generate_observation(idx=0, action_type=action_type, query=query)
+    obs, external_tool_info = strategy.generate_observation(
+        idx=0, action_type=action_type, query=query
+    )
     assert obs == gt_obs
     assert strategy._answer == "print('Hello World')"
     assert strategy._finished is False
@@ -144,7 +148,9 @@ def test_generate_observation() -> None:
     query = "def first_repeated_char(s):\n    char_set = set()\n    for char in s:\n        if char in char_set:\n            return char\n        else:\n            char_set.add(char)\n    return None"
     llm = FakeListChatModel(responses=[])
     strategy = ReActCodeStrategy(llm=llm)
-    obs, external_tool_info = strategy.generate_observation(idx=0, action_type=action_type, query=query)
+    obs, external_tool_info = strategy.generate_observation(
+        idx=0, action_type=action_type, query=query
+    )
     assert obs == gt_obs
     assert strategy._answer == query
     assert strategy._finished is True
@@ -157,7 +163,9 @@ def test_generate_observation() -> None:
     query = "def first_repeated_char(s):\n    char_set = set()\n    for char in s:\n        if char in char_set:\n            return char\n        else:\n            char_set.add(char)\n    return None"
     llm = FakeListChatModel(responses=[])
     strategy = ReActCodeStrategy(llm=llm)
-    obs, external_tool_info = strategy.generate_observation(idx=0, action_type=action_type, query=query)
+    obs, external_tool_info = strategy.generate_observation(
+        idx=0, action_type=action_type, query=query
+    )
     assert (
         obs
         == "Invalid Action. Valid Actions are Implement[code] Test[code] and Finish[answer]."
@@ -185,10 +193,12 @@ def test_create_output_dict() -> None:
         "query": query,
         "observation": obs,
         "answer": strategy._answer,
-        "external_tool_info": external_tool_info
+        "external_tool_info": external_tool_info,
     }
 
-    output = strategy.create_output_dict(thought, action_type, query, obs, external_tool_info)
+    output = strategy.create_output_dict(
+        thought, action_type, query, obs, external_tool_info
+    )
     assert output == expected_output