clustering.py

from sklearn.cluster import KMeans
from sklearn.metrics import silhouette_score
from sklearn.preprocessing import StandardScaler
import streamlit as st
import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd
from sklearn.decomposition import PCA

def summarize_cluster_characteristics(clustered_data, labels, cluster_number):
    cluster_data = clustered_data[labels == cluster_number]
    summary = cluster_data.mean().to_dict()
    return summary

def perform_clustering(df, n_clusters):
    df = df.dropna()

    scaler = StandardScaler()
    df_value_scaled = scaler.fit_transform(df)

    # Apply KMeans with the selected number of clusters
    model = KMeans(n_clusters=n_clusters, random_state=42)
    model.fit(df_value_scaled)
    labels = model.predict(df_value_scaled)
    score = silhouette_score(df_value_scaled, labels)

    df['Cluster'] = labels
    return df, score, df_value_scaled, labels, model

def plot_clusters(df_value_scaled, labels, new_data_point=None):
    pca = PCA(n_components=2)
    components = pca.fit_transform(df_value_scaled)
    df_components = pd.DataFrame(data=components, columns=['PC1', 'PC2'])
    df_components['Cluster'] = labels

    plt.figure(figsize=(10, 6))
    sns.scatterplot(x='PC1', y='PC2', hue='Cluster', data=df_components, palette='viridis', s=100, alpha=0.7)

    # Plot new data point if provided
    if new_data_point is not None:
        plt.scatter(new_data_point[:, 0], new_data_point[:, 1], color='red', marker='o', s=100, label='New Data Point')

    plt.title('Cluster Visualization')
    plt.xlabel('Principal Component 1')
    plt.ylabel('Principal Component 2')
    plt.legend(title='Cluster')
    st.pyplot(plt)