- Rishabh Agarwal
- Sahil Dhull
- Vipul Singhal
git clone https://github.com/SahilDhull/emphasis_selection.git
cd emphasis_selection
virtualenv project
pip install -r requirements.txt
Note: If not having virtualenv, install it using: pip install virtualenv
.
Download pre-trained GloVe embeddings from http://nlp.stanford.edu/data/glove.6B.zip
The problem statement (Task 10 in SemEval-2020) is designing automatic methods for emphasis selection i.e. choosing candidates for emphasis in short written text.
More formally, Given a sequence of words or tokens C = , we want to compute a score for each which indicates the degree of emphasis to be laid on the word.
-
datasets/ -> contains the train and development sets
-
BiLSTM_Attention_approach/ -> contains the model for 'BiLSTM + Attention' approach
-
Transformers_approach/ -> contains model and ensemble files for Transformers approach
-
requirements.txt -> contains versions of packages required
-
eval_metric.py -> contains the function for Match_m as defined in task