Matheel: A CLI and Python package for source-code similarity detection.
Project description
This is the repository for the demonstration paper "Matheel: A Hybrid Source Code Plagiarism Detection Software".
Matheel
Matheel is a Python package designed to detect source code similarity. It integrates semantic similarity models with traditional edit distance, providing a robust approach to detecting similarities among source code snippets.
Features
- Semantic Similarity: Uses Pre-Trained models.
- Edit-distance Metrics: Integrates Levenshtein and Jaro-Winkler similarity scores.
- Combined Weighted Similarity: Adjustable weights for semantic and syntactic similarity.
- Easy CLI & Python API: Suitable for both interactive and automated workflows.
- Interactive UI: Gradio-based user interface.
Installation
Install via pip:
pip install matheel
Usage
CLI Usage
Compare files within a compressed ZIP archive:
matheel compare codes.zip --model buelfhood/unixcoder-base-unimodal-ST --threshold 0.5 --num 100
Python API Usage
To calculate similarities programmatically:
from matheel.similarity import get_sim_list
# Define parameters
zip_file = "sample_codes.zip"
Ws, Wl, Wj = 0.7, 0.2, 0.1
model_name = "buelfhood/unixcoder-base-unimodal-ST"
threshold = 0.5
number_results = 100
# Get similarity results
results = get_sim_list(zip_file, Ws, Wl, Wj, model_name, threshold, number_results)
# Display results
print(results)
Gradio GUI:
The gradio_app folder contains a notebook that allows you to run the Gradio through a Jupyter Notebook. Also, a demo is available hosted on Huggingface Spaces.
Using Gradio API:
The tool can be used through the Gradio API as per the following call:
#pip install gradio_client
from gradio_client import Client, handle_file
client = Client("buelfhood/Matheel")
result = client.predict(
zipped_file=handle_file('zip file path'),
Ws=0.7,
Wl=0.3,
Wj=0,
model_name="buelfhood/unixcoder-base-unimodal-ST",
threshold=0,
number_results=10,
api_name="/get_sim_list"
)
print(result)
License
This project is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.
Acknowledgement:
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file matheel-0.1.8.tar.gz.
File metadata
- Download URL: matheel-0.1.8.tar.gz
- Upload date:
- Size: 10.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.12.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
eeee48eff393876372ce18acc4704ba6e8da72016647cb9c11cd7a639d999820
|
|
| MD5 |
a8d551254493f807bfa4b9276661619e
|
|
| BLAKE2b-256 |
a80dad2b9615a349ba7cb587777fa8876b029eb7d7066b1641b0eeec7da31743
|
File details
Details for the file matheel-0.1.8-py3-none-any.whl.
File metadata
- Download URL: matheel-0.1.8-py3-none-any.whl
- Upload date:
- Size: 10.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.12.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
2a73e76626da59a2d01c15fdf49fac9d11eb29328407e759558464c9105f6704
|
|
| MD5 |
481b6efcffd23f057e8f6231681f19dc
|
|
| BLAKE2b-256 |
880c544ce1ee77645712b83e449009aadee21b2268cf0a766dd74835a9ebfc56
|