I’ve been trying for the last 24 hours and the models neither load on the normal huggingface website nor when I try to deploy them as a specific inference endpoint. Could there be an issue with the model config itself?
Here are the logs from when I tried to deploy them on the dedicated inference api:
2024/05/04 22:02:02 ~ /usr/local/lib/python3.10/dist-packages/diffusers/utils/outputs.py:63: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.
2024/05/04 22:02:02 ~ torch.utils._pytree._register_pytree_node(
2024/05/04 22:02:02 ~ 2024-05-04 16:32:02,510 | INFO | Initializing model from directory:/repository
2024/05/04 22:02:02 ~ 2024-05-04 16:32:02,510 | INFO | No custom pipeline found at /repository/handler.py
2024/05/04 22:02:02 ~ 2024-05-04 16:32:02,510 | INFO | Using device CPU
2024/05/04 22:02:02 ~ Traceback (most recent call last):
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/starlette/routing.py”, line 732, in lifespan
2024/05/04 22:02:02 ~ async with self.lifespan_context(app) as maybe_state:
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/starlette/routing.py”, line 608, in aenter
2024/05/04 22:02:02 ~ await self._router.startup()
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/starlette/routing.py”, line 709, in startup
2024/05/04 22:02:02 ~ await handler()
2024/05/04 22:02:02 ~ File “/app/webservice_starlette.py”, line 60, in some_startup_task
2024/05/04 22:02:02 ~ inference_handler = get_inference_handler_either_custom_or_default_handler(HF_MODEL_DIR, task=HF_TASK)
2024/05/04 22:02:02 ~ File “/app/huggingface_inference_toolkit/handler.py”, line 54, in get_inference_handler_either_custom_or_default_handler
2024/05/04 22:02:02 ~ return HuggingFaceHandler(model_dir=model_dir, task=task)
2024/05/04 22:02:02 ~ File “/app/huggingface_inference_toolkit/handler.py”, line 18, in init
2024/05/04 22:02:02 ~ self.pipeline = get_pipeline(
2024/05/04 22:02:02 ~ File “/app/huggingface_inference_toolkit/utils.py”, line 276, in get_pipeline
2024/05/04 22:02:02 ~ hf_pipeline = pipeline(
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/transformers/pipelines/init.py”, line 815, in pipeline
2024/05/04 22:02:02 ~ config = AutoConfig.from_pretrained(
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/transformers/models/auto/configuration_auto.py”, line 1111, in from_pretrained
2024/05/04 22:02:02 ~ config_dict, unused_kwargs = PretrainedConfig.get_config_dict(pretrained_model_name_or_path, **kwargs)
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/transformers/configuration_utils.py”, line 633, in get_config_dict
2024/05/04 22:02:02 ~ config_dict, kwargs = cls._get_config_dict(pretrained_model_name_or_path, **kwargs)
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/transformers/configuration_utils.py”, line 688, in _get_config_dict
2024/05/04 22:02:02 ~ resolved_config_file = cached_file(
2024/05/04 22:02:02 ~ File “/usr/local/lib/python3.10/dist-packages/transformers/utils/hub.py”, line 369, in cached_file
2024/05/04 22:02:02 ~ raise EnvironmentError(
2024/05/04 22:02:02 ~ OSError: /repository does not appear to have a file named config.json. Checkout ‘https://huggingface.co//repository/None’ for available files.
- 2024-05-04T16:32:02.512+00:00
2024/05/04 22:02:02 ~ Application startup failed. Exiting.