TypeError: argument of type 'NoneType' is not iterable

Traceback (most recent call last):
File "/home/gaoshida/STVG/VideoLLaMA3/videollama3/infer.py", line 63, in
main()
File "/home/gaoshida/STVG/VideoLLaMA3/videollama3/infer.py", line 43, in main
model, processor = model_init(model_path)
File "/home/gaoshida/STVG/VideoLLaMA3/./videollama3/__init__.py", line 28, in model_init
tokenizer, model, image_processor, context_len = load_pretrained_model(model_path, None, model_name, **kwargs)
File "/home/gaoshida/STVG/VideoLLaMA3/./videollama3/model/__init__.py", line 147, in load_pretrained_model
model = Videollama3Qwen2ForCausalLM.from_pretrained(model_path, low_cpu_mem_usage=True, config=config, **kwargs)
File "/home/gaoshida/anaconda3/envs/videollama3/lib/python3.10/site-packages/transformers/modeling_utils.py", line 4097, in from_pretrained
model = cls(config, *model_args, **model_kwargs)
File "/home/gaoshida/STVG/VideoLLaMA3/./videollama3/model/videollama3_qwen2.py", line 50, in __init__
self.model = Videollama3Qwen2Model(config)
File "/home/gaoshida/STVG/VideoLLaMA3/./videollama3/model/videollama3_qwen2.py", line 42, in __init__
super(Videollama3Qwen2Model, self).__init__(config)
File "/home/gaoshida/STVG/VideoLLaMA3/./videollama3/model/videollama3_arch.py", line 59, in __init__
self.vision_encoder = build_vision_encoder(config, delay_load=False)
File "/home/gaoshida/STVG/VideoLLaMA3/./videollama3/model/encoder.py", line 273, in build_vision_encoder
if 'clip' in vision_encoder:
TypeError: argument of type 'NoneType' is not iterable

When running infer.py, an error TypeError: argument of type 'NoneType' is not iterable is reported