learning-loop-node 0.18.0__tar.gz → 0.18.2__tar.gz

{learning_loop_node-0.18.0 → learning_loop_node-0.18.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: learning-loop-node
-Version: 0.18.0
+Version: 0.18.2
 Summary: Python Library for Nodes which connect to the Zauberzeug Learning Loop
 Home-page: https://github.com/zauberzeug/learning_loop_node
 License: MIT

{learning_loop_node-0.18.0 → learning_loop_node-0.18.2}/learning_loop_node/detector/detector_logic.py

@@ -21,6 +21,12 @@ class DetectorLogic():
         self.model_info = None
 
     def load_model_info_and_init_model(self):
+        """
+        Load model information from disk and initialize the model.
+
+        The detector node uses a lock to make sure that this is not called 
+        concurrently with evaluate() or batch_evaluate(). 
+        """
         logging.info('Loading model from %s', GLOBALS.data_folder)
         self.model_info = ModelInformation.load_from_disk(f'{GLOBALS.data_folder}/model')
         if self.model_info is None:
@@ -42,21 +48,32 @@ class DetectorLogic():
 
     @abstractmethod
     def init(self):
-        """Called when a (new) model was loaded. Initialize the model. Model information available via `self.model_info`"""
+        """
+        Initialize the model.
+
+        Called when a (new) model was loaded. 
+        Model information available via `self.model_info`
+        The detector node uses a lock to make sure that this is not called
+        concurrently with evaluate() or batch_evaluate(). 
+        """
 
     @abstractmethod
     def evaluate(self, image: np.ndarray) -> ImageMetadata:
-        """Evaluate the image and return the detections.
+        """
+        Evaluate the image and return the detections.
 
         Called by the detector node when an image should be evaluated (REST or SocketIO).
         The resulting detections should be stored in the ImageMetadata.
         Tags stored in the ImageMetadata will be uploaded to the learning loop.
-        The function should return empty metadata if the detector is not initialized."""
+        The function should return empty metadata if the detector is not initialized.
+        """
 
     @abstractmethod
     def batch_evaluate(self, images: List[np.ndarray]) -> ImagesMetadata:
-        """Evaluate a batch of images and return the detections.
+        """
+        Evaluate a batch of images and return the detections.
 
         The resulting detections per image should be stored in the ImagesMetadata.
         Tags stored in the ImagesMetadata will be uploaded to the learning loop.
-        The function should return empty metadata if the detector is not initialized."""
+        The function should return empty metadata if the detector is not initialized.
+        """

{learning_loop_node-0.18.0 → learning_loop_node-0.18.2}/learning_loop_node/detector/detector_node.py

@@ -190,13 +190,13 @@ class DetectorNode(Node):
 
         # simulate startup
         await self.detector_logic.soft_reload()
-        self.detector_logic.load_model_info_and_init_model()
+        await self._load_model_info_and_init_model()
         self.operation_mode = OperationMode.Idle
 
     async def on_startup(self) -> None:
         try:
             self.outbox.ensure_continuous_upload()
-            self.detector_logic.load_model_info_and_init_model()
+            await self._load_model_info_and_init_model()
         except Exception:
             self.log.exception("error during 'startup'")
         self.operation_mode = OperationMode.Idle
@@ -217,7 +217,7 @@ class DetectorNode(Node):
         # Initialize the Socket.IO server with 20MB buffer size
         self.sio = socketio.AsyncServer(
             async_mode='asgi',
-            max_http_buffer_size=2e7,  # 20MB
+            max_http_buffer_size=64 * 1024 * 1024,  # 64 MiB
         )
         # Initialize and mount the ASGI app
         self.sio_app = socketio.ASGIApp(self.sio, socketio_path='/socket.io')
@@ -522,7 +522,7 @@ class DetectorNode(Node):
             self.log.info('Updated symlink for model to %s', os.readlink(model_symlink))
 
             try:
-                self.detector_logic.load_model_info_and_init_model()
+                await self._load_model_info_and_init_model()
             except NodeNeedsRestartError:
                 self.log.error('Node needs restart')
                 sys.exit(0)
@@ -535,6 +535,10 @@ class DetectorNode(Node):
             await self._sync_status_with_loop()
             # self.reload(reason='new model installed')
 
+    async def _load_model_info_and_init_model(self) -> None:
+        async with self.detection_lock:
+            self.detector_logic.load_model_info_and_init_model()
+
 # ================================== API Implementations ==================================
 
     async def set_operation_mode(self, mode: OperationMode):
@@ -570,11 +574,8 @@ class DetectorNode(Node):
         cares about uploading to the loop and returns the detections as ImageMetadata object.
         """
 
-        await self.detection_lock.acquire()
-        try:
+        async with self.detection_lock:
             metadata = await run.io_bound(self.detector_logic.evaluate, image)
-        finally:
-            self.detection_lock.release()
 
         metadata.tags.extend(tags)
         metadata.source = source
@@ -606,14 +607,12 @@ class DetectorNode(Node):
         """
         Processing function for the detector node when a a batch inference is requested via SocketIO.
 
-        This function infers the detections from all images, cares about uploading to the loop and returns the detections as a list of ImageMetadata.
+        This function infers the detections from all images, 
+        cares about uploading to the loop and returns the detections as a list of ImageMetadata.
         """
 
-        await self.detection_lock.acquire()
-        try:
+        async with self.detection_lock:
             all_detections = await run.io_bound(self.detector_logic.batch_evaluate, images)
-        finally:
-            self.detection_lock.release()
 
         for metadata in all_detections.items:
             metadata.tags.extend(tags)

{learning_loop_node-0.18.0 → learning_loop_node-0.18.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "learning_loop_node"
-version = "v0.18.0"
+version = "v0.18.2"
 description = "Python Library for Nodes which connect to the Zauberzeug Learning Loop"
 authors = ["Zauberzeug GmbH <info@zauberzeug.com>"]
 license = "MIT"