Traceback (most recent call last):
File "/home/acer/FlagAI/examples/vit_cifar100/train_deepspeed.py", line 69, in <module>
loader = AutoLoader(task_name="classification",
File "/home/acer/FlagAI/flagai/auto_model/auto_loader.py", line 333, in __init__
self.model = getattr(LazyImport(self.model_name[0]),
File "/home/acer/FlagAI/flagai/model/base_model.py", line 190, in from_pretrain
return load_local(checkpoint_path, only_download_config=only_download_config)
File "/home/acer/FlagAI/flagai/model/base_model.py", line 118, in load_local
model = cls.init_from_json(config_path, device=device, **kwargs)
File "/home/acer/FlagAI/flagai/model/base_model.py", line 75, in init_from_json
model = cls(change_json_to_cls(args), **kwargs)
File "/home/acer/FlagAI/flagai/model/vision/vit.py", line 232, in __init__
vit_config = VitConfig(**config)
TypeError: flagai.model.vision.vit.VitConfig() argument after ** must be a mapping, not ConfigObj
Information
[X] The official example scripts
[ ] My own modified scripts
Tasks
[X] An officially supported task in the examples folder (such as T5/AltCLIP, ...)
System Info
Information
Tasks
examples
folder (such as T5/AltCLIP, ...)Reproduction
examples/vit_cifar100/train_deepspeed.py
Expected behavior
tranining success