ihc-mdapy 0.1.0

Maximum Dissimilarity Algorithm in Python

mdapy

Introduction

Welcome to the "mdapy" repository! This repository contains the implementation of the Maximum Dissimilarity Algorithm in Python. The algorithm is commonly used in clustering and sampling to select the most dissimilar data points from a set. This README provides an overview of the project, a theoretical explanation of the algorithm, and examples of how to execute the code.

The Maximum Dissimilarity Algorithm

The Maximum Dissimilarity Algorithm is a technique used to select a specified number of data points from a larger dataset in such a way that the selected points are maximally dissimilar to each other. The algorithm is particularly useful in scenarios where you need to create diverse subsets of data for analysis or for initializing cluster centroids in clustering algorithms.

In this case, directional variables can also be considered, in which case, the distance between two angles is calculated as the shortest distance between the two angles in a circle. For example, the distance between 350° and 10° is 20°, not 340°. This is the approach used in this implementation of the algorithm (see Guanche et al. (2013) for more details).

Algorithm Steps

1. Initialization: Start with an empty subset, S. Choose the first data point arbitrarily or based on some criteria and add it to S.

2. Iteration: Repeat the following steps until you have selected the desired number of data points (n_clusters).

3. Finding the Most Dissimilar Point: For each data point not in S, calculate the minimum distance from that point to the points in S.

4. Selecting the Maximum Dissimilarity Point: Choose the data point with the greatest dissimilarity (maximum distance calculated in the previous step) and include it in subset S.

5. Output: Return the subset S, which contains n_clusters data points that are maximally dissimilar to each other.

How to Execute the Code

You can use the Maximum Dissimilarity Algorithm by following these steps:

1. Clone this repository to your local machine:

   git clone https://github.com/yourusername/mdapy.git
   

2. In the same directory, execute the following command to install the package:

   pip install .
   

3. Import the max_diss_alg function from the mdapy package:

   from mdapy import max_diss_alg
   

4. Create a tuple with the data arrays to use (must have same lengths) and choose the number of data points to select:

   tp = [14.93, 10.42, 7.46, 7.63, 12.20, 12.05, 14.49, 8.62]
   hs = [1.82, 1.45, 0.26, 0.90, 1.30, 5.71, 1.40, 2.18]
   dr = [268, 272, 289, 293, 276, 273, 276, 292]
   data = (tp, hs, dr)
   n_clusters = 3
   

5. Aditionally, choose the seed point to start the algorithm (optional) and, in case any of the variables is direcciotal, provide the position on the tuple of that array.

   seed_index = np.argmax(tp)
   dir_indices = [2]
   

6. Call the max_diss_alg function and store the output data:

   clustered_data = max_diss_alg(data, n_clusters, seed_index, dir_indices)
   

References

[1] Guanche, Y., Guanche, R., Camus, P., Mendez, F. J., & Medina, R. (2013). A multivariate approach to estimate design loads for offshore wind turbines. Wind Energy, 16(7), 1091-1106. (Apen)