MattePalte/thinking-like-a-developer

Companion Repository for "Thinking Like a Developer? Comparing the Attention of Humans with Neural Models of Code", accepted paper at ASE '21

28
/ 100
Experimental

This project helps software analytics and engineering researchers understand how humans interpret code compared to AI models. It provides preprocessed human attention data on code inspection and precomputed AI model attention weights. Researchers can input their own model's explainability metrics and compare them against human data to assess the model's strengths and weaknesses.

No commits in the last 6 months.

Use this if you are developing or evaluating a software analytics model and want to benchmark its explainability by comparing its 'attention' or feature importance to how human developers actually inspect code.

Not ideal if you are looking for a general-purpose code analysis tool or if you are not involved in academic software engineering research comparing human and AI code comprehension.

software-analytics developer-cognition human-computer-interaction code-comprehension explainable-ai-for-code
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 4 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

8

Forks

1

Language

Jupyter Notebook

License

MIT

Last pushed

Jul 14, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/MattePalte/thinking-like-a-developer"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.