from flask import Flask, render_template
import pandas as pd
from sklearn.manifold import TSNE
from sklearn.preprocessing import LabelEncoder
from sklearn.cluster import KMeans
import matplotlib.pyplot as plt
from io import BytesIO
import base64

app = Flask(__name__)

@app.route('/')
def index():
    # Load the dataset
    df = pd.read_csv('top_240_restaurants_recommended_in_los_angeles_2.csv')

    # Select the columns for clustering
    columns_for_clustering = ['Rank', 'StarRating', 'NumberOfReviews', 'Style']

    # Encode the 'Style' column
    label_encoder = LabelEncoder()
    df['Style_encoded'] = label_encoder.fit_transform(df['Style'])

    # Select the columns for clustering
    data_for_clustering = df[['Rank', 'StarRating', 'NumberOfReviews', 'Style_encoded']]

    # Perform t-SNE for dimensionality reduction
    tsne = TSNE(n_components=2, random_state=42)
    tsne_result = tsne.fit_transform(data_for_clustering)

    # Perform k-means clustering
    kmeans = KMeans(n_clusters=3, random_state=42)
    df['Cluster'] = kmeans.fit_predict(data_for_clustering)

    # Add t-SNE results to the DataFrame
    df['tsne_1'] = tsne_result[:, 0]
    df['tsne_2'] = tsne_result[:, 1]

    # Save the figure to a BytesIO object
    plt.scatter(df['tsne_1'], df['tsne_2'], c=df['Cluster'], cmap='viridis')
    plt.title('t-SNE Clustering')
    plt.xlabel('t-SNE Component 1')
    plt.ylabel('t-SNE Component 2')
    plt.tight_layout()
    img_data = BytesIO()
    plt.savefig(img_data, format='png')
    img_data.seek(0)

    # Convert the BytesIO object to base64 for embedding in HTML
    img_base64 = base64.b64encode(img_data.getvalue()).decode('utf-8')

    # Render the HTML template with the embedded image
    return render_template('index.html', img_base64=img_base64)

if __name__ == '__main__':
    app.run(debug=True)