search

JuanFMontesinos / Acappella-YNet

Official implementation of A cappella: Audio-visual Singing VoiceSeparation, from BMVC21
https://ipcv.github.io/Acappella/
14 stars 4 forks source link

Missing required arguments ? #2

Closed Enescigdem closed 2 years ago

Enescigdem commented 2 years ago

Hello, thanks for this great work and dataset, When I tried to run ,I got error below, Traceback (most recent call last): File "Desktop/Acappella-YNet/run.py", line 27, in iter_param, model, model_kwargs = VnBSS.ModelConstructor( File "/Desktop/Acappella-YNet/VnBSS/models/init.py", line 134, in build return self._build_dev() File "/Desktop/Acappella-YNet/VnBSS/models/init.py", line 140, in _build_dev model = constructor(**self.common_kwargs) TypeError: init() missing 6 required keyword-only arguments: 'remix_input', 'remix_coef', 'video_enabled', 'llcp_enabled', 'skeleton_enabled', and 'activation'

JuanFMontesinos commented 2 years ago

Hi, May I know which command did you use to run the code?

Enescigdem commented 2 years ago

Hi, This is exactly te same format in the README. python run.py -m model_name --workname experiment_name --arxiv_path directory_of_experiments --pretrained_from path_pret_weights

Enescigdem commented 2 years ago

Could you please tell me the correct way for training from scratch ?

JuanFMontesinos commented 2 years ago

Hi, should be fixed at f05d89e We are basically wondering how to give an easy acess to the dataset as you cannot run the code without it. Whether doing a mock up dataset so that code can be debugged or just uploading processed files.

I accidentally deleted few lines while removing irrelevant pieces of code.

Enescigdem commented 2 years ago

I cloned this GitHub repo 2 days ago so I must be using the newest version. I preprocessed the first 200 data files by using preproc/preprocess.py. Then tried to run just to see the pipeline of the system. So sorry I did not get your point . What is wrong with my trial?

JuanFMontesinos commented 2 years ago

Sorry i commited this after your issue. So you should clone it again. I just basically meant the code will stuck if you have no dataset.

We are gonna upload a mock up dataset within few hours so that the code can be debugged without having a dataset. If you clone the new code, please remember to set the dataset paths at config>dataset_paths.json

Enescigdem commented 2 years ago

Should I RE-run the preprocess.py ? Thank you for fast reply,

JuanFMontesinos commented 2 years ago

Yes, you can double check the dataset structure at https://github.com/JuanFMontesinos/Acappella-YNet/tree/main/VnBSS

Enescigdem commented 2 years ago

Thanks to your help . I could run the preprocessing without error.