fix: reset counter after each loop

ref: N25B-395

.env.example

@@ -3,9 +3,6 @@
 # The hostname of the Robot Interface. Change if the Control Backend and Robot Interface are running on different computers.
 RI_HOST="localhost"
 
-# The hostname of the User Interface. This is what the browser displays in the URL bar. Strangely, even if the UI is running on a different host than the backend, if the computer with the browser is also hosting the UI itself, this value should be http://localhost.
-UI_HOST="http://localhost:5173"
-
 # URL for the local LLM API. Must be an API that implements the OpenAI Chat Completions API, but most do.
 LLM_SETTINGS__LOCAL_LLM_URL="http://localhost:1234/v1/chat/completions"
 
@@ -15,8 +12,8 @@ LLM_SETTINGS__LOCAL_LLM_MODEL="gpt-oss"
 # Number of non-speech chunks to wait before speech ended. A chunk is approximately 31 ms. Increasing this number allows longer pauses in speech, but also increases response time.
 BEHAVIOUR_SETTINGS__VAD_NON_SPEECH_PATIENCE_CHUNKS=15
 
-# Timeout in milliseconds for socket polling. Increase this number if network latency/jitter is high, often the case when using Wi-Fi. Perhaps 500 ms or more. A symptom of this issue is transcriptions getting cut off.
-BEHAVIOUR_SETTINGS__SOCKET_POLLER_TIMEOUT_MS=400
+# Timeout in milliseconds for socket polling. Increase this number if network latency/jitter is high, often the case when using Wi-Fi. Perhaps 500 ms. A symptom of this issue is transcriptions getting cut off.
+BEHAVIOUR_SETTINGS__SOCKET_POLLER_TIMEOUT_MS=100
 
 
 

pyproject.toml

@@ -24,7 +24,6 @@ dependencies = [
   "sphinx-rtd-theme>=3.0.2",
   "tf-keras>=2.20.1",
   "torch>=2.8.0",
-  "tornado ; sys_platform == 'win32'",
   "uvicorn>=0.37.0",
 ]
 

src/control_backend/agents/bdi/agentspeak_generator.py

@@ -4,7 +4,6 @@ University within the Software Project course.
 © Copyright Utrecht University (Department of Information and Computing Sciences)
 """
 
-import logging
 from functools import singledispatchmethod
 
 from slugify import slugify
@@ -31,6 +30,7 @@ from control_backend.schemas.program import (
     BasicNorm,
     ConditionalNorm,
     EmotionBelief,
+    FaceBelief,
     GestureAction,
     Goal,
     InferredBelief,
@@ -67,7 +67,6 @@ class AgentSpeakGenerator:
     """
 
     _asp: AstProgram
-    logger = logging.getLogger(__name__)
 
     def generate(self, program: Program) -> str:
         """
@@ -107,7 +106,7 @@ class AgentSpeakGenerator:
         check if a keyword is a substring of the user's message.
 
         The generated rule has the form:
-        keyword_said(Keyword) :- user_said(Message) & .substring_case_insensitive(Keyword, Message, Pos) & Pos >= 0
+        keyword_said(Keyword) :- user_said(Message) & .substring(Keyword, Message, Pos) & Pos >= 0
 
         This enables the system to trigger behaviors based on keyword detection.
         """
@@ -119,7 +118,7 @@ class AgentSpeakGenerator:
             AstRule(
                 AstLiteral("keyword_said", [keyword]),
                 AstLiteral("user_said", [message])
-                & AstLiteral(".substring_case_insensitive", [keyword, message, position])
+                & AstLiteral(".substring", [keyword, message, position])
                 & (position >= 0),
             )
         )
@@ -135,6 +134,7 @@ class AgentSpeakGenerator:
         """
         self._add_reply_with_goal_plan()
         self._add_say_plan()
+        self._add_reply_plan()
         self._add_notify_cycle_plan()
 
     def _add_reply_with_goal_plan(self):
@@ -198,6 +198,40 @@ class AgentSpeakGenerator:
             )
         )
 
+    def _add_reply_plan(self):
+        """
+        Adds a plan for general reply actions.
+
+        This plan handles general reply actions where the agent needs to respond
+        to user input without a specific conversational goal. It:
+        1. Marks that the agent has responded this turn
+        2. Gathers all active norms
+        3. Generates a reply based on the user message and norms
+
+        Trigger: +!reply
+        Context: user_said(Message)
+        """
+        self._asp.plans.append(
+            AstPlan(
+                TriggerType.ADDED_GOAL,
+                AstLiteral("reply"),
+                [AstLiteral("user_said", [AstVar("Message")])],
+                [
+                    AstStatement(StatementType.ADD_BELIEF, AstLiteral("responded_this_turn")),
+                    AstStatement(
+                        StatementType.DO_ACTION,
+                        AstLiteral(
+                            "findall",
+                            [AstVar("Norm"), AstLiteral("norm", [AstVar("Norm")]), AstVar("Norms")],
+                        ),
+                    ),
+                    AstStatement(
+                        StatementType.DO_ACTION,
+                        AstLiteral("reply", [AstVar("Message"), AstVar("Norms")]),
+                    ),
+                ],
+            )
+        )
 
     def _add_notify_cycle_plan(self):
         """
@@ -235,39 +269,6 @@ class AgentSpeakGenerator:
             )
         )
 
-    def _add_stop_plan(self, phase: Phase):
-        """
-        Adds a plan to stop the program. This just skips to the end phase,
-        where there is no behavior defined.
-        """
-        self._asp.plans.append(
-            AstPlan(
-                TriggerType.ADDED_GOAL,
-                AstLiteral("stop"),
-                [AstLiteral("phase", [AstString(phase.id)])],
-                [
-                    AstStatement(
-                        StatementType.DO_ACTION,
-                        AstLiteral(
-                            "notify_transition_phase",
-                            [
-                                AstString(phase.id),
-                                AstString("end")
-                            ]
-                        )
-                    ),
-                    AstStatement(
-                        StatementType.REMOVE_BELIEF,
-                        AstLiteral("phase", [AstVar("Phase")]),
-                    ),
-                    AstStatement(
-                        StatementType.ADD_BELIEF,
-                        AstLiteral("phase", [AstString("end")])
-                    )
-                ]
-            )
-        )
-
     def _process_phases(self, phases: list[Phase]) -> None:
         """
         Processes all phases in the program and their transitions.
@@ -284,6 +285,21 @@ class AgentSpeakGenerator:
                 self._process_phase(curr_phase)
             self._add_phase_transition(curr_phase, next_phase)
 
+        # End phase behavior
+        # When deleting this, the entire `reply` plan and action can be deleted
+        self._asp.plans.append(
+            AstPlan(
+                type=TriggerType.ADDED_BELIEF,
+                trigger_literal=AstLiteral("user_said", [AstVar("Message")]),
+                context=[AstLiteral("phase", [AstString("end")])],
+                body=[
+                    AstStatement(
+                        StatementType.DO_ACTION, AstLiteral("notify_user_said", [AstVar("Message")])
+                    ),
+                    AstStatement(StatementType.ACHIEVE_GOAL, AstLiteral("reply")),
+                ],
+            )
+        )
 
     def _process_phase(self, phase: Phase) -> None:
         """
@@ -310,9 +326,6 @@ class AgentSpeakGenerator:
         for trigger in phase.triggers:
             self._process_trigger(trigger, phase)
 
-        # Add force transition to end phase
-        self._add_stop_plan(phase)
-
     def _add_phase_transition(self, from_phase: Phase | None, to_phase: Phase | None) -> None:
         """
         Adds plans for transitioning between phases.
@@ -488,13 +501,9 @@ class AgentSpeakGenerator:
             if isinstance(step, Goal):
                 subgoals.append(step)
 
-        if not goal.can_fail:
+        if not goal.can_fail and not continues_response:
             body.append(AstStatement(StatementType.ADD_BELIEF, self._astify(goal, achieved=True)))
 
-        if len(body) == 0:
-            self.logger.warning("Goal with no plan detected: %s", goal.name)
-            body.append(AstStatement(StatementType.EMPTY, AstLiteral("true")))
-
         self._asp.plans.append(AstPlan(TriggerType.ADDED_GOAL, self._astify(goal), context, body))
 
         self._asp.plans.append(
@@ -555,10 +564,10 @@ class AgentSpeakGenerator:
             )
         )
         for step in trigger.plan.steps:
-            if isinstance(step, Goal):
-                new_step = step.model_copy(update={"can_fail": False})  # triggers are sequence
-                subgoals.append(new_step)
             body.append(self._step_to_statement(step))
+            if isinstance(step, Goal):
+                step.can_fail = False  # triggers are continuous sequence
+                subgoals.append(step)
 
         # Arbitrary wait for UI to display nicely
         body.append(
@@ -602,7 +611,6 @@ class AgentSpeakGenerator:
         - check_triggers: When no triggers are applicable
         - transition_phase: When phase transition conditions aren't met
         - force_transition_phase: When forced transitions aren't possible
-        - stop: When we are already in the end phase
         """
         # Trigger fallback
         self._asp.plans.append(
@@ -634,16 +642,6 @@ class AgentSpeakGenerator:
             )
         )
 
-        # Stop fallback
-        self._asp.plans.append(
-            AstPlan(
-                TriggerType.ADDED_GOAL,
-                AstLiteral("stop"),
-                [],
-                [AstStatement(StatementType.EMPTY, AstLiteral("true"))],
-            )
-        )
-
     @singledispatchmethod
     def _astify(self, element: ProgramElement) -> AstExpression:
         """
@@ -690,6 +688,10 @@ class AgentSpeakGenerator:
     def _(self, eb: EmotionBelief) -> AstExpression:
         return AstLiteral("emotion_detected", [AstAtom(eb.emotion)])
 
+    @_astify.register
+    def _(self, fb: FaceBelief) -> AstExpression:
+        return AstLiteral("face_present")
+
     @_astify.register
     def _(self, ib: InferredBelief) -> AstExpression:
         """

src/control_backend/agents/bdi/bdi_core_agent.py

@@ -176,8 +176,6 @@ class BDICoreAgent(BaseAgent):
                         self._force_norm(msg.body)
                     case "force_next_phase":
                         self._force_next_phase()
-                    case "stop":
-                        self._stop()
                     case _:
                         self.logger.warning("Received unknown user interruption: %s", msg)
 
@@ -337,11 +335,6 @@ class BDICoreAgent(BaseAgent):
 
         self.logger.info("Manually forced phase transition.")
 
-    def _stop(self):
-        self._set_goal("stop")
-
-        self.logger.info("Stopped the program (skipped to end phase).")
-
     def _add_custom_actions(self) -> None:
         """
         Add any custom actions here. Inside `@self.actions.add()`, the first argument is
@@ -349,28 +342,6 @@ class BDICoreAgent(BaseAgent):
         the function expects (which will be located in `term.args`).
         """
 
-        @self.actions.add(".substring_case_insensitive", 3)
-        @agentspeak.optimizer.function_like
-        def _substring(agent, term, intention):
-            """
-            Find out if a string is a substring of another (case insensitive). Copied mostly from
-            the agentspeak library method .substring.
-            """
-            needle = agentspeak.asl_str(agentspeak.grounded(term.args[0], intention.scope)).lower()
-            haystack = agentspeak.asl_str(agentspeak.grounded(term.args[1], intention.scope)).lower()
-
-            choicepoint = object()
-
-            pos = haystack.find(needle)
-            while pos != -1:
-                intention.stack.append(choicepoint)
-
-                if agentspeak.unify(term.args[2], pos, intention.scope, intention.stack):
-                    yield
-
-                agentspeak.reroll(intention.scope, intention.stack, choicepoint)
-                pos = haystack.find(needle, pos + 1)
-
         @self.actions.add(".reply", 2)
         def _reply(agent, term, intention):
             """
@@ -496,6 +467,7 @@ class BDICoreAgent(BaseAgent):
                 body=str(trigger_name),
             )
 
+            # TODO: check with Pim
             self.add_behavior(self.send(msg))
 
             yield

src/control_backend/agents/bdi/text_belief_extractor_agent.py

@@ -538,9 +538,10 @@ class GoalAchievementInferrer(SemanticBeliefInferrer):
     async def _infer_goal(self, conversation: ChatHistory, goal: BaseGoal) -> bool:
         prompt = f"""{self._format_conversation(conversation)}
 
-Given the above conversation, has the following goal been achieved?
+Given the above conversation, what has the following goal been achieved?
 
-Description of the goal: {goal.description or goal.name}
+The name of the goal: {goal.name}
+Description of the goal: {goal.description}
 
 Answer with literally only `true` or `false` (without backticks)."""
 

src/control_backend/agents/perception/vad_agent.py

@@ -241,20 +241,9 @@ class VADAgent(BaseAgent):
                 self._reset_needed = False
 
             assert self.audio_in_poller is not None
-
-            non_speech_patience = settings.behaviour_settings.vad_non_speech_patience_chunks
-            begin_silence_length = settings.behaviour_settings.vad_begin_silence_chunks
-            prob_threshold = settings.behaviour_settings.vad_prob_threshold
-
             data = await self.audio_in_poller.poll()
             if data is None:
                 if len(self.audio_buffer) > 0:
-                    # Failed to receive new audio. Send remaining buffer to be transcribed.
-                    if len(self.audio_buffer) > begin_silence_length * 512:
-                        self.logger.debug("Speech ended.")
-                        assert self.audio_out_socket is not None
-                        await self.audio_out_socket.send(self.audio_buffer[: -2 * 512].tobytes())
-                    else:
                     self.logger.debug(
                         "No audio data received. Discarding buffer until new data arrives."
                     )
@@ -266,6 +255,9 @@ class VADAgent(BaseAgent):
             chunk = np.frombuffer(data, dtype=np.float32).copy()
             assert self.model is not None
             prob = self.model(torch.from_numpy(chunk), settings.vad_settings.sample_rate_hz).item()
+            non_speech_patience = settings.behaviour_settings.vad_non_speech_patience_chunks
+            begin_silence_length = settings.behaviour_settings.vad_begin_silence_chunks
+            prob_threshold = settings.behaviour_settings.vad_prob_threshold
 
             if prob > prob_threshold:
                 if self.i_since_speech > non_speech_patience + begin_silence_length:

src/control_backend/agents/perception/visual_emotion_recognition_agent/visual_emotion_recognition_agent.py

@@ -14,7 +14,7 @@ from control_backend.agents.perception.visual_emotion_recognition_agent.visual_e
 )
 from control_backend.core.agent_system import InternalMessage
 from control_backend.core.config import settings
-from control_backend.schemas.belief_message import Belief
+from control_backend.schemas.belief_message import Belief, BeliefMessage
 
 
 class VisualEmotionRecognitionAgent(BaseAgent):
@@ -44,6 +44,7 @@ class VisualEmotionRecognitionAgent(BaseAgent):
         self.timeout_ms = timeout_ms
         self.window_duration = window_duration
         self.min_frames_required = min_frames_required
+        self._face_detected = False
 
         # Pause functionality
         # NOTE: flag is set when running, cleared when paused
@@ -89,6 +90,9 @@ class VisualEmotionRecognitionAgent(BaseAgent):
         # Tracks counts of detected emotions per face index
         face_stats = defaultdict(Counter)
 
+        # How many times a face has been detected
+        face_detection_yes_no = [0, 0]
+
         prev_dominant_emotions = set()
 
         while self._running:
@@ -97,8 +101,8 @@ class VisualEmotionRecognitionAgent(BaseAgent):
 
                 width, height, image_bytes = await self.video_in_socket.recv_multipart()
 
-                width = int.from_bytes(width, 'little')
-                height = int.from_bytes(height, 'little')
+                width = int.from_bytes(width, "little")
+                height = int.from_bytes(height, "little")
 
                 # Convert bytes to a numpy buffer
                 image_array = np.frombuffer(image_bytes, np.uint8)
@@ -107,6 +111,13 @@ class VisualEmotionRecognitionAgent(BaseAgent):
 
                 # Get the dominant emotion from each face
                 current_emotions = self.emotion_recognizer.sorted_dominant_emotions(frame)
+
+                # Update face face_detection_yes_no
+                if len(current_emotions) > 0:
+                    face_detection_yes_no[0] += 1
+                else:
+                    face_detection_yes_no[1] += 1
+
                 # Update emotion counts for each detected face
                 for i, emotion in enumerate(current_emotions):
                     face_stats[i][emotion] += 1
@@ -122,18 +133,31 @@ class VisualEmotionRecognitionAgent(BaseAgent):
                             dominant_emotion = counter.most_common(1)[0][0]
                             window_dominant_emotions.add(dominant_emotion)
 
+                    if (
+                        face_detection_yes_no[0] > face_detection_yes_no[1]
+                        and not self._face_detected
+                    ):
+                        self._face_detected = True
+                        await self._inform_face_detected()
+                    elif (
+                        face_detection_yes_no[0] <= face_detection_yes_no[1] and self._face_detected
+                    ):
+                        self._face_detected = False
+                        await self._inform_face_detected()
+
+                    face_detection_yes_no = [0, 0]
+
                     await self.update_emotions(prev_dominant_emotions, window_dominant_emotions)
                     prev_dominant_emotions = window_dominant_emotions
                     face_stats.clear()
                     next_window_time = time.time() + self.window_duration
 
             except zmq.Again:
-                pass
+                self.logger.warning("No video frame received within timeout.")
 
             except Exception as e:
                 self.logger.error(f"Error in emotion recognition loop: {e}")
 
-    
     async def update_emotions(self, prev_emotions: set[str], emotions: set[str]):
         """
         Compare emotions from previous window and current emotions,
@@ -149,9 +173,7 @@ class VisualEmotionRecognitionAgent(BaseAgent):
         for emotion in emotions_to_remove:
             self.logger.info(f"Emotion '{emotion}' has disappeared.")
             try:
-                emotion_beliefs_remove.append(
-                    Belief(name="emotion_detected", arguments=[emotion], remove=True)
-                )
+                emotion_beliefs_remove.append(Belief(name="emotion_detected", arguments=[emotion]))
             except ValidationError:
                 self.logger.warning("Invalid belief for emotion removal: %s", emotion)
 
@@ -175,6 +197,20 @@ class VisualEmotionRecognitionAgent(BaseAgent):
         )
         await self.send(message)
 
+    async def _inform_face_detected(self):
+        if self._face_detected:
+            belief_message = BeliefMessage(create=[Belief(name="face_present")])
+        else:
+            belief_message = BeliefMessage(delete=[Belief(name="face_present")])
+
+        msg = InternalMessage(
+            to=settings.agent_settings.bdi_core_name,
+            thread="beliefs",
+            body=belief_message.model_dump_json(),
+        )
+
+        await self.send(msg)
+
     async def handle_message(self, msg: InternalMessage):
         """
         Handle incoming messages.
@@ -204,4 +240,3 @@ class VisualEmotionRecognitionAgent(BaseAgent):
         """
         self.video_in_socket.close()
         await super().stop()
-        

src/control_backend/agents/user_interrupt/user_interrupt_agent.py

@@ -164,12 +164,6 @@ class UserInterruptAgent(BaseAgent):
                     else:
                         self.logger.info("Sent resume command.")
 
-                case "stop":
-                    self.logger.debug(
-                        "Received stop command."
-                    )
-                    await self._send_stop_command()
-
                 case "next_phase" | "reset_phase":
                     await self._send_experiment_control_to_bdi_core(event_type)
                 case _:
@@ -429,15 +423,3 @@ class UserInterruptAgent(BaseAgent):
             await self.send(vad_message)
             # Voice Activity Detection and Visual Emotion Recognition agents
             self.logger.info("Sent resume command to VAD and VED agents.")
-
-    async def _send_stop_command(self):
-        """
-        Send a command to the BDI to stop the program (i.e., skip to end phase).
-        """
-        msg = InternalMessage(
-            to=settings.agent_settings.bdi_core_name,
-            body="",
-            thread="stop"
-        )
-
-        await self.send(msg)

src/control_backend/api/v1/endpoints/robot.py

@@ -123,7 +123,7 @@ async def ping_stream(request: Request):
         sub_socket.setsockopt(zmq.SUBSCRIBE, b"ping")
         connected = False
 
-        ping_frequency = settings.behaviour_settings.sleep_s + 1
+        ping_frequency = 2
 
         # Even though its most likely the updates should alternate
         # (So, True - False - True - False for connectivity),

src/control_backend/core/config.py

@@ -112,7 +112,7 @@ class BehaviourSettings(BaseModel):
     conversation_history_length_limit: int = 10
 
     # Visual Emotion Recognition settings
-    visual_emotion_recognition_window_duration_s: int = 5
+    visual_emotion_recognition_window_duration_s: int = 3
     visual_emotion_recognition_min_frames_per_face: int = 3
     # AgentSpeak related settings
     trigger_time_to_wait: int = 2000

src/control_backend/schemas/program.py

@@ -7,7 +7,7 @@ University within the Software Project course.
 from enum import Enum
 from typing import Literal
 
-from pydantic import UUID4, BaseModel, field_validator
+from pydantic import UUID4, BaseModel
 
 
 class ProgramElement(BaseModel):
@@ -24,13 +24,6 @@ class ProgramElement(BaseModel):
     # To make program elements hashable
     model_config = {"frozen": True}
 
-    @field_validator("name")
-    @classmethod
-    def name_must_not_start_with_number(cls, v: str) -> str:
-        if v and v[0].isdigit():
-            raise ValueError('Field "name" must not start with a number.')
-        return v
-
 
 class LogicalOperator(Enum):
     """
@@ -48,8 +41,8 @@ class LogicalOperator(Enum):
     OR = "OR"
 
 
-type Belief = KeywordBelief | SemanticBelief | InferredBelief | EmotionBelief
-type BasicBelief = KeywordBelief | SemanticBelief | EmotionBelief
+type Belief = KeywordBelief | SemanticBelief | InferredBelief | EmotionBelief | FaceBelief
+type BasicBelief = KeywordBelief | SemanticBelief | EmotionBelief | FaceBelief
 
 
 class KeywordBelief(ProgramElement):
@@ -124,6 +117,15 @@ class EmotionBelief(ProgramElement):
     emotion: str
 
 
+class FaceBelief(ProgramElement):
+    """
+    Represents the belief that at least one face is currently in view.
+    """
+
+    name: str = ""
+    face_present: bool
+
+
 class Norm(ProgramElement):
     """
     Base class for behavioral norms that guide the robot's interactions.

uv.lock

@@ -1,5 +1,5 @@
 version = 1
-revision = 2
+revision = 3
 requires-python = ">=3.13"
 resolution-markers = [
     "python_full_version >= '3.14' and sys_platform == 'darwin'",
@@ -1524,7 +1524,6 @@ dependencies = [
     { name = "sphinx-rtd-theme" },
     { name = "tf-keras" },
     { name = "torch" },
-    { name = "tornado", marker = "sys_platform == 'win32'" },
     { name = "uvicorn" },
 ]
 
@@ -1580,7 +1579,6 @@ requires-dist = [
     { name = "sphinx-rtd-theme", specifier = ">=3.0.2" },
     { name = "tf-keras", specifier = ">=2.20.1" },
     { name = "torch", specifier = ">=2.8.0" },
-    { name = "tornado", marker = "sys_platform == 'win32'" },
     { name = "uvicorn", specifier = ">=0.37.0" },
 ]
 
@@ -2726,17 +2724,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/52/27/7fc2d7435af044ffbe0b9b8e98d99eac096d43f128a5cde23c04825d5dcf/torchaudio-2.8.0-cp313-cp313t-win_amd64.whl", hash = "sha256:d4a715d09ac28c920d031ee1e60ecbc91e8a5079ad8c61c0277e658436c821a6", size = 2549553, upload-time = "2025-08-06T14:59:00.019Z" },
 ]
 
-[[package]]
-name = "tornado"
-version = "6.5.4"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/37/1d/0a336abf618272d53f62ebe274f712e213f5a03c0b2339575430b8362ef2/tornado-6.5.4.tar.gz", hash = "sha256:a22fa9047405d03260b483980635f0b041989d8bcc9a313f8fe18b411d84b1d7", size = 513632, upload-time = "2025-12-15T19:21:03.836Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/0c/1a/d7592328d037d36f2d2462f4bc1fbb383eec9278bc786c1b111cbbd44cfa/tornado-6.5.4-cp39-abi3-win32.whl", hash = "sha256:1768110f2411d5cd281bac0a090f707223ce77fd110424361092859e089b38d1", size = 446481, upload-time = "2025-12-15T19:21:00.008Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/6d/c69be695a0a64fd37a97db12355a035a6d90f79067a3cf936ec2b1dc38cd/tornado-6.5.4-cp39-abi3-win_amd64.whl", hash = "sha256:fa07d31e0cd85c60713f2b995da613588aa03e1303d75705dca6af8babc18ddc", size = 446886, upload-time = "2025-12-15T19:21:01.287Z" },
-    { url = "https://files.pythonhosted.org/packages/50/49/8dc3fd90902f70084bd2cd059d576ddb4f8bb44c2c7c0e33a11422acb17e/tornado-6.5.4-cp39-abi3-win_arm64.whl", hash = "sha256:053e6e16701eb6cbe641f308f4c1a9541f91b6261991160391bfc342e8a551a1", size = 445910, upload-time = "2025-12-15T19:21:02.571Z" },
-]
-
 [[package]]
 name = "tqdm"
 version = "4.67.1"