charlie-the-coding-cow-classroom

Charlie the Coding Cow: Classroom Edition

https://github.com/arjunguha/charlie-the-coding-cow-classroom

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

✓
CITATION.cff file
Found CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
○
DOI references
○
Academic publication links
○
Academic email domains
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (11.4%) to scientific vocabulary

Last synced: 10 months ago · JSON representation ·

Repository

Charlie the Coding Cow: Classroom Edition

Basic Info

Host: GitHub
Owner: arjunguha
License: bsd-3-clause
Language: Python
Default Branch: main
Size: 487 KB

Statistics

Stars: 2
Watchers: 1
Forks: 0
Open Issues: 0
Releases: 0

Created almost 2 years ago · Last pushed over 1 year ago

Metadata Files

Readme License Citation

Charlie the Coding Cow - Classroom Edition

Charlie the Coding Cow Screenshot

This is a simplified version of the "Charlie the Coding Cow" experiment platform, designed for classroom use. The original version of this platform was used in the following papers:

Hannah Babe, Sydney Nguyen, Yangtian Zi, Arjun Guha, Molly Q Feldman, and Carolyn Jane Anderson. StudentEval: a Benchmark of Student-Written Prompts for Large Language Models of Code. ACL Findings 2024.
Molly Q Feldman and Carolyn Jane Anderson. Non-Expert Programmers in the Generative AI Future. CHIWORK 2024.
Sydney Nguyen, Hannah Babe, Yangtian Zi, Arjun Guha, Carolyn Jane Anderson, and Molly Q Feldman. How Beginning Programmers and Code LLMs (Mis)read Each Other. CHI 2024.

This version features a single sequence of problems for all users (defined in tasks.jsonl), without end-of-problem surveys or a built-in tutorial (which can be run separately in class). It simplifies data storage by logging all results to results_USERNAME.jsonl files. For authentication, it just uses a list of usernames without any passwords (defined in users.txt)

The file all_tasks.jsonl contains all the tasks from the original platform, and is derived from problems.yaml.

Requirements

Python 3
pip3 install gradio openai
An LLM endpoint supporting the OpenAI completions API

Usage

Set environment variables:
- BASE_URL
- API_KEY
Run the application: python3 app.py --model MODEL_NAME
For more options: python3 app.py --help

Example

The following command starts the program with gpt-4o-mini, and makes it publicly accessible:

bash API_KEY=$OPENAI_API_KEY python3 app.py --model gpt-4o-mini --share --is-chat

After the program starts, Gradio will print the public URL to the console.

Owner

Name: Arjun Guha
Login: arjunguha
Kind: user
Location: Boston, MA
Company: @nuprl

Website: https://khoury.northeastern.edu/~arjunguha/
Twitter: arjunguha
Repositories: 29
Profile: https://github.com/arjunguha

hacker / cs professor

Citation (CITATION.cff)

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
  - family-names: "Guha"
    given-names: "Arjun"
date-released: 2024-10-07
url: "https://github.com/arjunguha/charlie-the-coding-cow-classroom"
title: "Charlie the Coding Cow: Classroom Edition"
version: 1.0
preferred-citation:
  type: conference-paper
  authors:
    - family-names: "Nguyen"
      given-names: "Sydney"
    - family-names: "Babe"
      given-names: "Hannah McLean"
    - family-names: "Zi"
      given-names: "Yangtian"
    - family-names: "Guha"
      given-names: "Arjun"
    - family-names: "Anderson"
      given-names: "Carolyn Jane"
    - family-names: "Feldman"
      given-names: "Molly Q"
  title: "How Beginning Programmers and Code LLMs (Mis)read Each Other"
  conference:
    name: "ACM Conference on Human Factors in Computing Systems"
    abbreviation: "CHI"
  year: 2024

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Open Source Science