EngineeringSoftware/time-segmented-evaluation

Code and data for "Impact of Evaluation Methodologies on Code Summarization" in ACL 2022.

28
/ 100
Experimental

This project helps researchers and practitioners in machine learning for code understand how different ways of splitting a dataset impact the results of code summarization models. It takes in raw code and comments with timestamps, processes them, and then trains and evaluates several machine learning models. The output shows how various evaluation methodologies affect model performance, helping users choose the most appropriate method for their research or application.

No commits in the last 6 months.

Use this if you are a researcher or ML engineer working on code summarization and want to rigorously test how different data splitting strategies (like time-segmented, mixed-project, or cross-project) influence your model's real-world applicability.

Not ideal if you are looking for a pre-trained code summarization model to use directly in an application without needing to perform deep methodological analysis.

code-summarization machine-learning-evaluation natural-language-processing-code software-engineering-research dataset-splitting
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

10

Forks

1

Language

Python

License

MIT

Last pushed

Sep 06, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/EngineeringSoftware/time-segmented-evaluation"

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