环境是PyTorch，无法运行，提示要下载[full]版本，tf下没有问题

HanLP
#1

之前是tf环境，运行正常，现在我卸载了tf，安装了PyTorch，代码无法运行，提示要[full]版本，我尝试安装它，结果它在下载tensorflow 2.6，这是怎么回事，我不想要tf

λ python .\test.py
Traceback (most recent call last):
  File ".\test.py", line 3, in <module>
    tokenizer = hanlp.load('PKU_NAME_MERGED_SIX_MONTHS_CONVSEG')
  File "d:\NLP\anaconda3\lib\site-packages\hanlp\__init__.py", line 43, in load
    return load_from_meta_file(save_dir, 'meta.json', verbose=verbose, **kwargs)
  File "d:\NLP\anaconda3\lib\site-packages\hanlp\utils\component_util.py", line 109, in load_from_meta_file
    '\n\n\tpip install hanlp[full]') from None
ModuleNotFoundError: Some modules required by this model are missing. Please install the full version:

        pip install hanlp[full]

请问hanlp如何选择使用什么引擎？比如两种环境同时存在的情况下？