jina-ai / clip-as-service

🏄 Scalable embedding, reasoning, ranking for images and sentences with CLIP

Home Page:https://clip-as-service.jina.ai

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

An exception when executing "python -m clip_server"

sawsam opened this issue · comments

INFO gateway/rep-0@32304 start server bound to 127.0.0.1:51000 [10/30/23 10:55:09]
UserWarning: docs annotation must be a type hint, got DocumentArray instead, you should maybe remove the string annotation. Default valueDocumentArray will be used instead. (raised from C:\Python311\Lib\site-packages\jina\serve\executors_init_.py:267)
CRITI… clip_t/rep-0@23436 can not load the executor from {"jtype": "CLIPEncoder", "metas": {"py_modules": ["clip_server.executors.clip_torch"]}} [10/30/23 10:55:12]
ERROR clip_t/rep-0@23436 TypeError("VisionTransformer.init() got an unexpected keyword argument 'global_average_pool'") during 'WorkerRuntime'
initialization
add "--quiet-error" to suppress the exception details
Traceback (most recent call last):
File "C:\Python311\Lib\site-packages\jina\serve\executors\run.py", line 140, in run
runtime = AsyncNewLoopRuntime(
^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\asyncio.py", line 92, in init
self._loop.run_until_complete(self.async_setup())
File "C:\Python311\Lib\asyncio\base_events.py", line 653, in run_until_complete
return future.result()
^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\asyncio.py", line 309, in async_setup
self.server = self._get_server()
^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\asyncio.py", line 214, in get_server
return GRPCServer(
^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\servers\grpc.py", line 34, in init
super().init(**kwargs)
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\servers_init
.py", line 63, in init
] = (req_handler or self.get_request_handler())
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\servers_init
.py", line 88, in get_request_handler
return self.req_handler_cls(
^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\worker\request_handling.py", line 139, in init
self.load_executor(
File "C:\Python311\Lib\site-packages\jina\serve\runtimes\worker\request_handling.py", line 374, in load_executor
self.executor: BaseExecutor = BaseExecutor.load_config(
^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\jaml_init
.py", line 792, in load_config
obj = JAML.load(tag_yml, substitute=False, runtime_args=runtime_args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\jaml_init
.py", line 174, in load
r = yaml.load(stream, Loader=get_jina_loader_with_runtime(runtime_args))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\yaml_init
.py", line 81, in load
return loader.get_single_data()
^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\yaml\constructor.py", line 51, in get_single_data
return self.construct_document(node)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\yaml\constructor.py", line 55, in construct_document
data = self.construct_object(node)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\yaml\constructor.py", line 100, in construct_object
data = constructor(self, node)
^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\jaml_init
.py", line 582, in _from_yaml
return get_parser(cls, version=data.get('version', None)).parse(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\jaml\parsers\executor\legacy.py", line 46, in parse
obj = cls(
^^^^
File "C:\Python311\Lib\site-packages\jina\serve\executors\decorators.py", line 58, in arg_wrapper
f = func(self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\serve\helper.py", line 73, in arg_wrapper
f = func(self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\clip_server\executors\clip_torch.py", line 91, in init
self._model = CLIPModel(
^^^^^^^^^^
File "C:\Python311\Lib\site-packages\clip_server\model\openclip_model.py", line 38, in init
self._model = load_openai_model(
^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\clip_server\model\model.py", line 477, in load_openai_model
model = build_model_from_openai_state_dict(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\clip_server\model\model.py", line 413, in build_model_from_openai_state_dict
model = CLIP(
^^^^^
File "C:\Python311\Lib\site-packages\clip_server\model\model.py", line 281, in init
self.visual = _build_vision_tower(
^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\clip_server\model\model.py", line 193, in _build_vision_tower
visual = VisionTransformer(
^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\clip_server\model\model.py", line 83, in init
super().init(
TypeError: VisionTransformer.init() got an unexpected keyword argument 'global_average_pool'
ERROR Flow@13032 An exception occurred: [10/30/23 10:55:12]
ERROR Flow@13032 Flow is aborted due to ['clip_t'] can not be started.
WARNI… gateway/rep-0@13032 Pod was forced to close after 1 second. Graceful closing is not available on Windows. [10/30/23 10:55:13]
Traceback (most recent call last):
File "", line 198, in _run_module_as_main
File "", line 88, in run_code
File "C:\Python311\Lib\site-packages\clip_server_main
.py", line 25, in
with f:
File "C:\Python311\Lib\site-packages\jina\orchestrate\orchestrator.py", line 14, in enter
return self.start()
^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\orchestrate\flow\builder.py", line 33, in arg_wrapper
return func(self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\site-packages\jina\orchestrate\flow\base.py", line 1846, in start
self._wait_until_all_ready()
File "C:\Python311\Lib\site-packages\jina\orchestrate\flow\base.py", line 2012, in _wait_until_all_ready
raise RuntimeFailToStart
jina.excepts.RuntimeFailToStart

Hi,

It seems that downgrading open-clip-torch allow to run clip_server

pip install open-clip-torch==2.22.0

Best Regards,
Michel.

it works, thank you very much!