First do the following to collect the dataset from our S3 bucket:

curl -OL http://craftassist.s3-us-west-2.amazonaws.com/pubr/collected_dataset.tar.gz -o collected_dataset.tar.gz
tar -xzvf collected_dataset.tar.gz -C collected_dataset/ --strip-components 1

This folder now contains the following :

• Grammar_Spec_doc.md a specificcation of our grammar and logical forms.
• annotation_tools/ : this folder contains the annotation tools with Turk support along with the postprocessing scripts for each.
• collected_dataset/ : this folder contains the interactive an dprompts datasets alogn witht he cross validaion splits we used for our experiments.
• model_training_code/ : this folder contains the training code for our models.
• writeup/ : This folder contains the writeup of the paper