CS 598 Midterm Project

Summary generation toolkit with configurable models.

Introduction

Thank you for going through this course project for CS598 - Learning in Humans and Machines. This readme file incorporates all the steps necessary to set up the application and get it up and running.

Authors

• @dhanur-sharma

Requirements

Please ensure that the following dependencies are installed before proceeding with the installation:

• Ubuntu 22.04.3 LTS
• Python 3.10
• git (git version 2.39.1)
• Pip 23.2

I'd recommend running it on ilab to ensure a smoother experience so all package dependencies are resolved.

Installation

Step 1: Unzip the folder

After navigating to the directory you'd like to install the application in, run the following command to unzip the folder:

sudo apt-get install unzip
unzip code.zip -d code
cd code

You're ready to run the application.

Step 2: Set configurations

Open the config.py file to access the parameters to change. Change the FILE_PATH to point to the data file 'hcV3-stories.csv' and the range of stories to be generated.

Step 3: Run the program

Run the following commands on ilab to ensure that the process is complete even if the SSH connection is interrupted:

keep-job 24
nohup python3 run.py

The output will be generated in the outputs directory and logs will be generated in nohup.out in the same folder.

Note: The API listed will only be valid as long as there are compute credits associated with this account.