readme

Author: ibeltagy

README.md

@@ -1,21 +1,26 @@
-# abstruct
-Adding structure to scientific abstracts
+# <p align=center>Sequential Sentence Classification</p>
+This repo has code and data for our paper "Pretrained Language Models for Sequential Sentence Classification"
 
-## Setting up
+### How to run
 
 ```
-# Create a Python environment with your environment mangager:
-
-conda create -n abstruct python=3.6
+pip install -r requirements.txt
+scripts/train.sh tmp_output_dir
+```
 
-source activate abstruct
+Update the `scripts/train.sh` script with the appropriate hyperparameters and datapaths.
 
-conda env create -f environment.yml
-```
+### CSAbstrcut dataset
 
-## Tests
+The train, dev, test splits of the dataset are in `data/CSAbstrcut`
+### Citing
 
+If you use the data or the model, please cite,
 ```
-pytest tests/
-```
-
+@inproceedings{Cohan2019EMNLP,
+  title={Pretrained Language Models for Sequential Sentence Classification},
+  author={Arman Cohan, Iz Beltagy, Daniel King, Bhavana Dalvi, Dan Weld},
+  year={2019},
+  booktitle={EMNLP},
+}
+```