Code Editor for app.py

import streamlit as st
import numpy as np
import pandas as pd
import plotly.express as px

st.title("3D Data Clustering Visualizer")

# Sidebar for user input
st.sidebar.header("User Input Parameters")
n_clusters = st.sidebar.slider("Number of clusters", 1, 10, 3)

# Generate random 3D data
np.random.seed(42)
X = np.random.rand(100, 3) * 100

# K-means clustering
from sklearn.cluster import KMeans
kmeans = KMeans(n_clusters=n_clusters, random_state=42)
clusters = kmeans.fit_predict(X)
centroids = kmeans.cluster_centers_

# Create a DataFrame
clusters_df = pd.DataFrame(X, columns=["X", "Y", "Z"])
clusters_df['Cluster'] = clusters
centroids_df = pd.DataFrame(centroids, columns=["X", "Y", "Z"])
centroids_df['Cluster'] = range(n_clusters)

# Create the 3D scatter plot
fig = px.scatter_3d(clusters_df, x='X', y='Y', z='Z', color='Cluster', title='3D Clustering Visualization',
                      labels={'X': 'X Axis', 'Y': 'Y Axis', 'Z': 'Z Axis'}, hover_name=clusters_df.index)

# Adding centroids to the plot
fig.add_trace(px.scatter_3d(centroids_df, x='X', y='Y', z='Z', color='Cluster',
                              color_discrete_sequence=['black'], marker=dict(size=10, symbol='x')).data[0])

st.plotly_chart(fig)