How do I resolve TypeError: Trainer.__init__() got an unexpected keyword argument 'resume_from_checkpoint'?

I'm not a developer.
I want to separate multiple speakers.
I tried everything from speechbrain to espnet, but it was too difficult, so I gave up.
I tried asteroid's wsj0-mix-var because it seemed easy, but I got the error in the title.
I'm not capable of fixing it, so I'm asking.
The error message is below.
Traceback (most recent call last):

File "C:\Users\Administrator\asteroid\egs\wsj0-mix-var\Multi-Decoder-DPRNN\train.py", line 219, in

main(arg_dic)

File "C:\Users\Administrator\asteroid\egs\wsj0-mix-var\Multi-Decoder-DPRNN\train.py", line 100, in main

trainer = pl.Trainer(

^^^^^^^^^^^

File "C:\ProgramData\anaconda3\envs\asteroid\Lib\site-packages\pytorch_lightning\utilities\argparse.py", lin
e 70, in insert_env_defaults

return fn(self, **kwargs)

^^^^^^^^^^^^^^^^^^

TypeError: Trainer.__init__() got an unexpected keyword argument 'resume_from_checkpoint'
We've created a custom dataset with preprocess_wsj0mix.py . Now we can start training with train.py .

Start at egs/wsj0-mix-var/Multi-Decoder-DPRNN/train.py line 100, where the traceback shows the pytorch-lightning Trainer is created, and inspect the surrounding training setup. Done means train.py can initialize Trainer and proceed to training without the reported TypeError.

Written by the indexing model from the issue text.