{
  "id": 606645,
  "title": "ANOTHER GATEWAY ERROR",
  "url": "/competitions/rsna-intracranial-aneurysm-detection/discussion/606645",
  "author_name": "NinjaNick",
  "post_date": "2025-09-09T13:17:22.488000",
  "votes": -1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Oh my god. So I finally get everything finished. But alas, my victories are not to be celebrated yet. I run my prediction script only to suffer from the local gateway just now working at all.</p>\n<pre><code> (&lt;GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: &gt;, \n</code></pre>\n<p>Here is my predict function:</p>\n<pre><code> () -&gt; pl.DataFrame | pd.DataFrame:\n    \n\n    series_id = os.path.basename(series_path)\n    ()\n\n    :\n        \n        volume = dicom_series_to_volume(series_path)\n\n         volume  :\n            ()\n            predictions_array = [] * (LABEL_COLS)\n        :\n            ()\n\n            \n            rgb_image = cuda6_volume_to_mip_rgb(volume, target_size=)\n            ()\n\n            \n            predictions_list = []\n\n            \n            image_tensor = inference_transform(rgb_image)\n            image_tensor = image_tensor.unsqueeze().to(device)\n\n             torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n\n            \n            rgb_flipped = np.fliplr(rgb_image.copy())\n            image_tensor = inference_transform(rgb_flipped)\n            image_tensor = image_tensor.unsqueeze().to(device)\n\n             torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n\n            \n            predictions_array = np.mean(predictions_list, axis=).tolist()\n\n            ()\n\n     Exception  e:\n        ()\n        predictions_array = [] * (LABEL_COLS)\n\n    \n    predictions = pl.DataFrame(\n        data=[[series_id] + predictions_array],\n        schema=[ID_COL] + LABEL_COLS,\n        orient=,\n    )\n\n    \n    shutil.rmtree(, ignore_errors=)\n\n     predictions.drop(ID_COL)\n\n\n\n()\n()\n()  \n()\n()\n\n\ninference_server = kaggle_evaluation.rsna_inference_server.RSNAInferenceServer(cuda6_predict)\n\n os.getenv():\n    ()\n    inference_server.serve()\n:\n    ()\n    inference_server.run_local_gateway()\n    display(pl.read_parquet())\n</code></pre>\n<p>but I keep getting this error:</p>\n<pre><code>---------------------------------------------------------------------------\nGatewayRuntimeError                       Traceback (most recent call last)\n/tmp/ipykernel_36/py  &lt;cell line: &gt;()\n     :\n         ()\n--&gt;      inference_server.run_local_gateway()\n         display(pl.read_parquet())\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py  run_local_gateway(, data_paths, file_share_dir, *args, **kwargs)\n                 .gateway.run()\n              Exception  err:\n--&gt;               err  \n             :\n                 .server.stop()\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py  run_local_gateway(, data_paths, file_share_dir, *args, **kwargs)\n             :\n                 .gateway = ._get_gateway_for_test(data_paths, file_share_dir, *args, **kwargs)\n--&gt;              .gateway.run()\n              Exception  err:\n                  err  \n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  run()\n              error:\n                 \n--&gt;               error\n     \n         @final\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  run()\n             :\n                 .unpack_data_paths()\n--&gt;              predictions, row_ids = .get_all_predictions()\n                 .write_submission(predictions, row_ids)\n              GatewayRuntimeError  gre:\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  get_all_predictions()\n             .data_batch_counter = \n              data_batch, row_ids  .generate_data_batches():\n--&gt;              predictions = .predict(*data_batch)\n                 .competition_agnostic_validation(predictions, row_ids)\n                 .competition_specific_validation(predictions, row_ids, data_batch)\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  predict(, *args, **kwargs)\n                  .client.send(, *args, **kwargs)\n              Exception  e:\n--&gt;              .handle_server_error(e, )\n     \n          () -&gt; :\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  handle_server_error(, exception, endpoint)\n                  GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_NEVER_STARTED)  \n                exception_str:\n--&gt;               GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT, )  \n                exception_str:\n                 \n\nGatewayRuntimeError: (&lt;GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: &gt;, )\n</code></pre>\n<p>This error is driving me insane. Is it just the local gateway or is there something else going on?</p>",
  "messages": [
    {
      "id": 3287665,
      "postDate": "2025-09-12T01:25:08.597Z",
      "content": "<p>Your predict function should be named predict(). </p>\n<p>It should match exactly of that in the demo submission. <br>\nThe server expects a function named predict. </p>",
      "rawMarkdown": "Your predict function should be named predict(). \n\nIt should match exactly of that in the demo submission. \nThe server expects a function named predict. ",
      "votes": 2
    },
    {
      "id": 3286242,
      "postDate": "2025-09-09T13:17:22.490Z",
      "content": "<p>Oh my god. So I finally get everything finished. But alas, my victories are not to be celebrated yet. I run my prediction script only to suffer from the local gateway just now working at all.</p>\n<pre><code> (&lt;GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: &gt;, \n</code></pre>\n<p>Here is my predict function:</p>\n<pre><code> () -&gt; pl.DataFrame | pd.DataFrame:\n    \n\n    series_id = os.path.basename(series_path)\n    ()\n\n    :\n        \n        volume = dicom_series_to_volume(series_path)\n\n         volume  :\n            ()\n            predictions_array = [] * (LABEL_COLS)\n        :\n            ()\n\n            \n            rgb_image = cuda6_volume_to_mip_rgb(volume, target_size=)\n            ()\n\n            \n            predictions_list = []\n\n            \n            image_tensor = inference_transform(rgb_image)\n            image_tensor = image_tensor.unsqueeze().to(device)\n\n             torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n\n            \n            rgb_flipped = np.fliplr(rgb_image.copy())\n            image_tensor = inference_transform(rgb_flipped)\n            image_tensor = image_tensor.unsqueeze().to(device)\n\n             torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n\n            \n            predictions_array = np.mean(predictions_list, axis=).tolist()\n\n            ()\n\n     Exception  e:\n        ()\n        predictions_array = [] * (LABEL_COLS)\n\n    \n    predictions = pl.DataFrame(\n        data=[[series_id] + predictions_array],\n        schema=[ID_COL] + LABEL_COLS,\n        orient=,\n    )\n\n    \n    shutil.rmtree(, ignore_errors=)\n\n     predictions.drop(ID_COL)\n\n\n\n()\n()\n()  \n()\n()\n\n\ninference_server = kaggle_evaluation.rsna_inference_server.RSNAInferenceServer(cuda6_predict)\n\n os.getenv():\n    ()\n    inference_server.serve()\n:\n    ()\n    inference_server.run_local_gateway()\n    display(pl.read_parquet())\n</code></pre>\n<p>but I keep getting this error:</p>\n<pre><code>---------------------------------------------------------------------------\nGatewayRuntimeError                       Traceback (most recent call last)\n/tmp/ipykernel_36/py  &lt;cell line: &gt;()\n     :\n         ()\n--&gt;      inference_server.run_local_gateway()\n         display(pl.read_parquet())\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py  run_local_gateway(, data_paths, file_share_dir, *args, **kwargs)\n                 .gateway.run()\n              Exception  err:\n--&gt;               err  \n             :\n                 .server.stop()\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py  run_local_gateway(, data_paths, file_share_dir, *args, **kwargs)\n             :\n                 .gateway = ._get_gateway_for_test(data_paths, file_share_dir, *args, **kwargs)\n--&gt;              .gateway.run()\n              Exception  err:\n                  err  \n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  run()\n              error:\n                 \n--&gt;               error\n     \n         @final\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  run()\n             :\n                 .unpack_data_paths()\n--&gt;              predictions, row_ids = .get_all_predictions()\n                 .write_submission(predictions, row_ids)\n              GatewayRuntimeError  gre:\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  get_all_predictions()\n             .data_batch_counter = \n              data_batch, row_ids  .generate_data_batches():\n--&gt;              predictions = .predict(*data_batch)\n                 .competition_agnostic_validation(predictions, row_ids)\n                 .competition_specific_validation(predictions, row_ids, data_batch)\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  predict(, *args, **kwargs)\n                  .client.send(, *args, **kwargs)\n              Exception  e:\n--&gt;              .handle_server_error(e, )\n     \n          () -&gt; :\n\n/kaggle//rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py  handle_server_error(, exception, endpoint)\n                  GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_NEVER_STARTED)  \n                exception_str:\n--&gt;               GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT, )  \n                exception_str:\n                 \n\nGatewayRuntimeError: (&lt;GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: &gt;, )\n</code></pre>\n<p>This error is driving me insane. Is it just the local gateway or is there something else going on?</p>",
      "rawMarkdown": "Oh my god. So I finally get everything finished. But alas, my victories are not to be celebrated yet. I run my prediction script only to suffer from the local gateway just now working at all.\n```\nGatewayRuntimeError: (<GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: 4>, 'Server did not register a listener for predict')\n```\n\nHere is my predict function:\n```\ndef cuda6_predict(series_path: str) -> pl.DataFrame | pd.DataFrame:\n    \"\"\"CUDA 6 compatible prediction function matching training preprocessing\"\"\"\n    \n    series_id = os.path.basename(series_path)\n    print(f\"Processing series: {series_id}\")\n    \n    try:\n        # Convert DICOM series to 3D volume\n        volume = dicom_series_to_volume(series_path)\n        \n        if volume is None:\n            print(\"Failed to load DICOM series, using default predictions\")\n            predictions_array = [0.5] * len(LABEL_COLS)\n        else:\n            print(f\"Loaded volume shape: {volume.shape}\")\n            \n            # CUDA 6 compatible preprocessing\n            rgb_image = cuda6_volume_to_mip_rgb(volume, target_size=224)\n            print(f\"Generated MIP RGB shape: {rgb_image.shape}\")\n            \n            # Test Time Augmentation (TTA) - simplified\n            predictions_list = []\n            \n            # Original image\n            image_tensor = inference_transform(rgb_image)\n            image_tensor = image_tensor.unsqueeze(0).to(device)\n            \n            with torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n            \n            # Horizontal flip\n            rgb_flipped = np.fliplr(rgb_image.copy())\n            image_tensor = inference_transform(rgb_flipped)\n            image_tensor = image_tensor.unsqueeze(0).to(device)\n            \n            with torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n            \n            # Average TTA predictions\n            predictions_array = np.mean(predictions_list, axis=0).tolist()\n            \n            print(f\"Generated TTA predictions: {predictions_array[:3]}... (showing first 3)\")\n    \n    except Exception as e:\n        print(f\"Error during prediction: {e}\")\n        predictions_array = [0.5] * len(LABEL_COLS)\n    \n    # Create DataFrame\n    predictions = pl.DataFrame(\n        data=[[series_id] + predictions_array],\n        schema=[ID_COL] + LABEL_COLS,\n        orient='row',\n    )\n    \n    # Clean up disk space\n    shutil.rmtree('/kaggle/shared', ignore_errors=True)\n    \n    return predictions.drop(ID_COL)\n\n\n# Debug info\nprint(\"SUBMISSION DEBUG INFO\")\nprint(f\"Current working directory: {os.getcwd()}\")\nprint(f\"Files in directory: {os.listdir('.')[:10]}\")  # First 10 files\nprint(f\"Model save path: {model_save_path}\")\nprint(f\"Device: {device}\")\n\n# Initialize inference server\ninference_server = kaggle_evaluation.rsna_inference_server.RSNAInferenceServer(cuda6_predict)\n\nif os.getenv('KAGGLE_IS_COMPETITION_RERUN'):\n    print(\"COMPETITION MODE - Starting inference server...\")\n    inference_server.serve()\nelse:\n    print(\"LOCAL MODE - Running test...\")\n    inference_server.run_local_gateway()\n    display(pl.read_parquet('/kaggle/working/submission.parquet'))\n```\n\nbut I keep getting this error:\n```\n---------------------------------------------------------------------------\nGatewayRuntimeError                       Traceback (most recent call last)\n/tmp/ipykernel_36/1387463453.py in <cell line: 0>()\n    396 else:\n    397     print(\"LOCAL MODE - Running test...\")\n--> 398     inference_server.run_local_gateway()\n    399     display(pl.read_parquet('/kaggle/working/submission.parquet'))\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py in run_local_gateway(self, data_paths, file_share_dir, *args, **kwargs)\n    108             self.gateway.run()\n    109         except Exception as err:\n--> 110             raise err from None\n    111         finally:\n    112             self.server.stop(0)\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py in run_local_gateway(self, data_paths, file_share_dir, *args, **kwargs)\n    106         try:\n    107             self.gateway = self._get_gateway_for_test(data_paths, file_share_dir, *args, **kwargs)\n--> 108             self.gateway.run()\n    109         except Exception as err:\n    110             raise err from None\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in run(self)\n    151         elif error:\n    152             # For local testing\n--> 153             raise error\n    154 \n    155     @final\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in run(self)\n    132         try:\n    133             self.unpack_data_paths()\n--> 134             predictions, row_ids = self.get_all_predictions()\n    135             self.write_submission(predictions, row_ids)\n    136         except GatewayRuntimeError as gre:\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in get_all_predictions(self)\n    108         self.data_batch_counter = 0\n    109         for data_batch, row_ids in self.generate_data_batches():\n--> 110             predictions = self.predict(*data_batch)\n    111             self.competition_agnostic_validation(predictions, row_ids)\n    112             self.competition_specific_validation(predictions, row_ids, data_batch)\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in predict(self, *args, **kwargs)\n    126             return self.client.send('predict', *args, **kwargs)\n    127         except Exception as e:\n--> 128             self.handle_server_error(e, 'predict')\n    129 \n    130     def run(self) -> None:\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in handle_server_error(self, exception, endpoint)\n    417             raise GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_NEVER_STARTED) from None\n    418         if f'No listener for {endpoint} was registered' in exception_str:\n--> 419             raise GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT, f'Server did not register a listener for {endpoint}') from None\n    420         if 'Exception calling application' in exception_str:\n    421             # Extract just the exception message raised by the inference server\n\nGatewayRuntimeError: (<GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: 4>, 'Server did not register a listener for predict')\n```\n\nThis error is driving me insane. Is it just the local gateway or is there something else going on?",
      "votes": -1
    }
  ],
  "comments": [
    {
      "id": 3287665,
      "author_name": "Parkhyeryn",
      "author_url": "",
      "post_date": "2025-09-12T01:25:08.597000",
      "content": "<p>Your predict function should be named predict(). </p>\n<p>It should match exactly of that in the demo submission. <br>\nThe server expects a function named predict. </p>",
      "votes": 2,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3287665": "Your predict function should be named predict(). \n\nIt should match exactly of that in the demo submission. \nThe server expects a function named predict. ",
    "3286242": "Oh my god. So I finally get everything finished. But alas, my victories are not to be celebrated yet. I run my prediction script only to suffer from the local gateway just now working at all.\n```\nGatewayRuntimeError: (<GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: 4>, 'Server did not register a listener for predict')\n```\n\nHere is my predict function:\n```\ndef cuda6_predict(series_path: str) -> pl.DataFrame | pd.DataFrame:\n    \"\"\"CUDA 6 compatible prediction function matching training preprocessing\"\"\"\n    \n    series_id = os.path.basename(series_path)\n    print(f\"Processing series: {series_id}\")\n    \n    try:\n        # Convert DICOM series to 3D volume\n        volume = dicom_series_to_volume(series_path)\n        \n        if volume is None:\n            print(\"Failed to load DICOM series, using default predictions\")\n            predictions_array = [0.5] * len(LABEL_COLS)\n        else:\n            print(f\"Loaded volume shape: {volume.shape}\")\n            \n            # CUDA 6 compatible preprocessing\n            rgb_image = cuda6_volume_to_mip_rgb(volume, target_size=224)\n            print(f\"Generated MIP RGB shape: {rgb_image.shape}\")\n            \n            # Test Time Augmentation (TTA) - simplified\n            predictions_list = []\n            \n            # Original image\n            image_tensor = inference_transform(rgb_image)\n            image_tensor = image_tensor.unsqueeze(0).to(device)\n            \n            with torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n            \n            # Horizontal flip\n            rgb_flipped = np.fliplr(rgb_image.copy())\n            image_tensor = inference_transform(rgb_flipped)\n            image_tensor = image_tensor.unsqueeze(0).to(device)\n            \n            with torch.no_grad():\n                logits = model(image_tensor)\n                probabilities = torch.sigmoid(logits).cpu().numpy().flatten()\n                predictions_list.append(probabilities)\n            \n            # Average TTA predictions\n            predictions_array = np.mean(predictions_list, axis=0).tolist()\n            \n            print(f\"Generated TTA predictions: {predictions_array[:3]}... (showing first 3)\")\n    \n    except Exception as e:\n        print(f\"Error during prediction: {e}\")\n        predictions_array = [0.5] * len(LABEL_COLS)\n    \n    # Create DataFrame\n    predictions = pl.DataFrame(\n        data=[[series_id] + predictions_array],\n        schema=[ID_COL] + LABEL_COLS,\n        orient='row',\n    )\n    \n    # Clean up disk space\n    shutil.rmtree('/kaggle/shared', ignore_errors=True)\n    \n    return predictions.drop(ID_COL)\n\n\n# Debug info\nprint(\"SUBMISSION DEBUG INFO\")\nprint(f\"Current working directory: {os.getcwd()}\")\nprint(f\"Files in directory: {os.listdir('.')[:10]}\")  # First 10 files\nprint(f\"Model save path: {model_save_path}\")\nprint(f\"Device: {device}\")\n\n# Initialize inference server\ninference_server = kaggle_evaluation.rsna_inference_server.RSNAInferenceServer(cuda6_predict)\n\nif os.getenv('KAGGLE_IS_COMPETITION_RERUN'):\n    print(\"COMPETITION MODE - Starting inference server...\")\n    inference_server.serve()\nelse:\n    print(\"LOCAL MODE - Running test...\")\n    inference_server.run_local_gateway()\n    display(pl.read_parquet('/kaggle/working/submission.parquet'))\n```\n\nbut I keep getting this error:\n```\n---------------------------------------------------------------------------\nGatewayRuntimeError                       Traceback (most recent call last)\n/tmp/ipykernel_36/1387463453.py in <cell line: 0>()\n    396 else:\n    397     print(\"LOCAL MODE - Running test...\")\n--> 398     inference_server.run_local_gateway()\n    399     display(pl.read_parquet('/kaggle/working/submission.parquet'))\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py in run_local_gateway(self, data_paths, file_share_dir, *args, **kwargs)\n    108             self.gateway.run()\n    109         except Exception as err:\n--> 110             raise err from None\n    111         finally:\n    112             self.server.stop(0)\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/templates.py in run_local_gateway(self, data_paths, file_share_dir, *args, **kwargs)\n    106         try:\n    107             self.gateway = self._get_gateway_for_test(data_paths, file_share_dir, *args, **kwargs)\n--> 108             self.gateway.run()\n    109         except Exception as err:\n    110             raise err from None\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in run(self)\n    151         elif error:\n    152             # For local testing\n--> 153             raise error\n    154 \n    155     @final\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in run(self)\n    132         try:\n    133             self.unpack_data_paths()\n--> 134             predictions, row_ids = self.get_all_predictions()\n    135             self.write_submission(predictions, row_ids)\n    136         except GatewayRuntimeError as gre:\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in get_all_predictions(self)\n    108         self.data_batch_counter = 0\n    109         for data_batch, row_ids in self.generate_data_batches():\n--> 110             predictions = self.predict(*data_batch)\n    111             self.competition_agnostic_validation(predictions, row_ids)\n    112             self.competition_specific_validation(predictions, row_ids, data_batch)\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in predict(self, *args, **kwargs)\n    126             return self.client.send('predict', *args, **kwargs)\n    127         except Exception as e:\n--> 128             self.handle_server_error(e, 'predict')\n    129 \n    130     def run(self) -> None:\n\n/kaggle/input/rsna-intracranial-aneurysm-detection/kaggle_evaluation/core/base_gateway.py in handle_server_error(self, exception, endpoint)\n    417             raise GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_NEVER_STARTED) from None\n    418         if f'No listener for {endpoint} was registered' in exception_str:\n--> 419             raise GatewayRuntimeError(GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT, f'Server did not register a listener for {endpoint}') from None\n    420         if 'Exception calling application' in exception_str:\n    421             # Extract just the exception message raised by the inference server\n\nGatewayRuntimeError: (<GatewayRuntimeErrorType.SERVER_MISSING_ENDPOINT: 4>, 'Server did not register a listener for predict')\n```\n\nThis error is driving me insane. Is it just the local gateway or is there something else going on?"
  }
}