Code Editor for app.py

import streamlit as st
import pandas as pd
import numpy as np
from sklearn.neighbors import NearestNeighbors
import io
from streamlit_plotly_events import plotly_events
import plotly.express as px

# Sample data creation for default testing
def create_sample_data():
    np.random.seed(42)
    n_samples = 20
    x = np.random.uniform(0, 10, n_samples)
    y = np.random.uniform(0, 10, n_samples)
    
    # Sample imgur URLs (these are just examples)
    image_urls = [
        "https://i.imgur.com/bY5XMel.png",
        "https://i.imgur.com/gHksBVP.png",
        "https://i.imgur.com/3vQUj5R.png",
        "https://i.imgur.com/zzIJpGC.png"
    ]
    
    # Randomly assign image URLs
    urls = np.random.choice(image_urls, n_samples)
    
    df = pd.DataFrame({
        'x': x,
        'y': y,
        'image_url': urls
    })
    return df

# Initialize session state for clicked point
if 'clicked_point' not in st.session_state:
    st.session_state.clicked_point = None

st.title("Interactive Scatter Plot with Images")

# File uploader
uploaded_file = st.file_uploader("Upload your CSV file (should contain x, y, and image_url columns)", type=['csv'])

# Load data
if uploaded_file is not None:
    df = pd.read_csv(uploaded_file)
else:
    st.write("Using sample data for demonstration. Upload your own CSV file to use your data.")
    df = create_sample_data()

# Display the dataframe
st.write("Dataset:")
st.dataframe(
    df,
    column_config={
        "image_url": st.column_config.ImageColumn(
            "Preview",
            help="Image preview"
        )
    }
)

# Create scatter plot
scatter_chart = px.scatter(
    df,
    x='x',
    y='y',
    height=400)

# Get clicked point from the chart
clicked_point = plotly_events(scatter_chart)

if clicked_point:
    x_clicked = clicked_point[0]['x']
    y_clicked = clicked_point[0]['y']
    
    # Find nearest neighbor
    X = df[['x', 'y']].values
    nbrs = NearestNeighbors(n_neighbors=1).fit(X)
    distances, indices = nbrs.kneighbors([[x_clicked, y_clicked]])
    
    nearest_point = df.iloc[indices[0][0]]
    
    # Display results
    col1, col2 = st.columns(2)
    
    with col1:
        st.write("Clicked Point:")
        st.write(f"X: {x_clicked:.2f}")
        st.write(f"Y: {y_clicked:.2f}")
        
    with col2:
        st.write("Nearest Point in Dataset:")
        st.write(f"X: {nearest_point['x']:.2f}")
        st.write(f"Y: {nearest_point['y']:.2f}")
    
    # Display the image
    st.write("Associated Image:")
    st.image(nearest_point['image_url'])