unsupervised_topic_segmentation icon indicating copy to clipboard operation
unsupervised_topic_segmentation copied to clipboard

example of input dataframe

Open Narabzad opened this issue 3 years ago • 3 comments

Thank you for the great work. Can you please bring one example of the inputs? i.e., what should be the data frame look like? more specifically what is caption_col_name ?

df: pd.DataFrame,
    meeting_id_col_name: str,
    start_col_name: str,
    end_col_name: str,
    caption_col_name: str,
    topic_segmentation_config: TopicSegmentationConfig,

Also, what are the hyperparameters you used for textiling_hyperparameters?

Narabzad avatar Jul 02 '21 06:07 Narabzad

This data structure is tailored to the datasets stored in our internal database. Therefore it's better to modify this dataframe according to your data format.

I pushed the hyperparameter values we had.

gdamaskinos avatar Jul 06 '21 11:07 gdamaskinos

Hello! Can you, please, share start_col_name and end_col_name columns data format for correct algorithm work? I`m DS intern and working on implementation your algorithm with Russian language.

BeniaDev avatar Aug 20 '21 10:08 BeniaDev

you can use this dataset https://github.com/Yale-LILY/QMSum

BMukhtar avatar Feb 24 '23 14:02 BMukhtar