ReMoDe is a Python library designed for the robust detection of modes in data distributions. It uses statistical tests, including Fisher's exact test and binomial tests, to determine if a given maximum in a data distribution is a true local maximum.
- Mode Detection: Identifies all potential local maxima in the dataset.
- Statistical Tests: Implements Fisher's exact test and binomial tests to validate modes.
- Data Formatting: Converts raw data into histogram format for analysis.
- Stability Analysis: Includes functionality to assess the stability of detected modes using jackknife resampling.
- Visualization: Provides methods to plot the histogram of data along with identified modes.
pip install git+https://git@github.com/sodascience/remode.git@mainHere is a simple example of how to use the ReMode library:
from remode import ReMoDe
# Sample data (histogram counts)
xt_count = [8, 20, 5, 2, 6, 2, 30]
# Create an instance of ReMoDe
detector = ReMoDe()
# Fit model
results = detector.fit(xt_count)
# Plot the results
detector.plot_maxima()
# Perform stability analysis
stability_info = detector.evaluate_stability(percentage_steps=50)Please cite the following paper
To be added
Contributions are what make the open source community an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.
Please refer to the CONTRIBUTING file for more information on issues and pull requests.
This project is licensed under the GNU GPLv3. This allows you to do almost anything they want with this project, except distributing closed source versions.
This project is a port of the R version of ReMoDe. It is maintained by the ODISSEI Social Data
Science (SoDa) team.
Do you have questions, suggestions, or remarks? File an issue in the issue tracker!