zehuiwu / MMML

Multi-Modality Multi-Loss Fusion Network
67 stars 9 forks source link

Hi,I'm a new comer of MSA.I want to start with your paper.But when I run run_audio.py,a KeyError: 'mode' occurred during runtime.Could you tell me how to fix it? #6

Open gongxuanchi opened 2 months ago

gongxuanchi commented 2 months ago

df = pd.read_csv(csv_path) df = df[df['mode']=='train'].reset_index()

error:KeyError: 'mode'

studentxxxsss commented 1 month ago

Hello blogger, mosi ran successfully, and then the same problem occurred when running the sims data set run.py. How to solve it? Is it the problem of csv file or format? In addition, when extracting the audio of the mosei dataset, part of the audio damage caused by the extraction of part of the folder is empty code can not run,How to solve it?

gongxuanchi commented 1 month ago

I'm in the same situation as you.I think it is the problem of the csv file.But I don't find any audio is damaged in my dataset.

studentxxxsss commented 1 month ago

When I was running the mosei dataset, some missing files popped up, I extracted the audio again, and it showed that some audio was being repaired, but the repair has not been completed yet and I do not know whether it will be successful in the future. May I ask if your data set was downloaded from the read me link, and my data set was downloaded from other places, I do not know whether this is the problem

gongxuanchi commented 1 month ago

I download from the README link

studentxxxsss commented 1 month ago

Ok, thanks for your reply, that should be a problem with the mosei dataset I downloaded