ishalyminov / multitask_disfluency_detection

Code for the paper "Multi-Task Learning for Domain-General Spoken Disfluency Detection in Dialogue Systems" (Igor Shalyminov, Arash Eshghi, and Oliver Lemon)
24 stars 2 forks source link

I have followed README,but there is no folder "swbd". So there may be several files missing #14

Closed lizhzh8 closed 4 years ago

lizhzh8 commented 4 years ago

python make_deep_disfluency_dataset.py swbd disfluency

image

ishalyminov commented 4 years ago

Hmm, are you sure you're using the code from the master branch without any modifications? This script only takes the required positional args: https://github.com/ishalyminov/multitask_disfluency_detection/blob/2148a04a63e592e277a6fd4131ec9b8287fb619a/make_deep_disfluency_dataset.py#L62

ishalyminov commented 4 years ago

To make it clear, when you invoke the script like this:

python make_deep_disfluency_dataset.py swbd disfluency

-- the script creates the swbd folder itself and puts there the processed dataset in the disfluency format. So no extra files or folders are required for it to run - apart from the deep_disfluency stuff which we covered in the previous issue.

The only problem I see with the output you have pasted is that --result_folder RESULT_FOLDER and --format FORMAT are both optional arguments whereas the original script takes 2 positional arguments as you can see at the above link. So please make sure you're using the version of the script exactly as at the link above.

lizhzh8 commented 4 years ago

Actually when I run git submodule update --init,there is some error.

image

ishalyminov commented 4 years ago

That looks weird, I experienced that as well recently. Mind creating another issue for that?

lizhzh8 commented 4 years ago

OK