k-means-app

Streamlit application to explore k-means clustering

Deployed with streamlit:

https://alicjakarczewska-k-means-app-kmeans-iris-qpq98y.streamlitapp.com/

The aim and use of application

This application shows the method k-means clustering on Iris dataset. It include 4 metrics (Euclidean, L1, Chebyshev and Mahalanobise distances) and various techniques for measuring similarity (accuracy score, jaccard score, silhouette sc, cosine similarity).

Used python libraries:

• streamlit,
• pandas,
• seaborn,
• sklearn,
• numpy,
• matplotlib...

Running aplication locally

It is recommended to firstly manage an environment, for example conda:

conda create --name <env_name> --file requirements.txt

And then activate it:

conda activate <env_name>

To run application use:

streamlit run kmeans-iris.py