This repository contains the code for all experiments described in the following paper:

Modeling Student Response Times: Towards Efficient One-on-one Tutoring Dialogues
Luciana Benotti, Jayadev Bhaskaran, Sigtryggur Kjartansson, David Lang
EMNLP Workshop on Noisy User Generated Text (W-NUT) at EMNLP 2018, Brussels, Belgium.

The paper can be found here.

1. Create a virtual environment called rt:

conda create -n rt python=3.6 anaconda

To enter the environment run

source activate rt

To exit the environment run

source deactivate rt

2. Enter the newly created virtual environment
3. Install package requirements

pip install -r requirements.txt

4. Install package

python setup.py install

Note: You will have to repeat steps 3. and 4. throughout development as we add new dependencies and modules.

• requirements.txt - make sure to update this file every time you add a new import!
• data/ - directory that will contain all data.
• runs/ - directory to store training run outputs.
• src/ - directory for all source code (e.g. models, trainers, analysis scripts etc.)
• src/config.py - central location for any project structure constants (e.g. paths to data). This is not intended for model parameters.