from flask import Flask, request, render_template
from sklearn.datasets import make_moons
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import PolynomialFeatures
from sklearn.neural_network import MLPRegressor
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt

app = Flask(__name__)

@app.route('/')
def home():
    return render_template('index.html')

@app.route('/compare_models', methods=['POST'])
def compare_models():
    # Генерация данных
    rs = 0
    X, y = make_moons(noise=0.3, random_state=rs)
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=rs)

    # Линейная регрессия
    lr = LinearRegression()
    lr.fit(X_train, y_train)
    lr_score = lr.score(X_test, y_test)

    # Полиномиальная регрессия (степень 3)
    poly = PolynomialFeatures(degree=3)
    X_poly = poly.fit_transform(X_train)
    poly_reg = LinearRegression()
    poly_reg.fit(X_poly, y_train)
    poly_score = poly_reg.score(poly.transform(X_test), y_test)

    # Многослойный персептрон (100 нейронов в скрытом слое, alpha=0.01)
    mlp = MLPRegressor(hidden_layer_sizes=(100,), alpha=0.01, random_state=rs)
    mlp.fit(X_train, y_train)
    mlp_score = mlp.score(X_test, y_test)

    # Создание графиков
    plt.figure(figsize=(15, 4))

    plt.subplot(131)
    plt.scatter(X_test[:, 0], X_test[:, 1], c=y_test, cmap=plt.cm.RdBu)
    plt.title('Линейная регрессия\n(Score: {:.2f})'.format(lr_score))

    plt.subplot(132)
    plt.scatter(X_test[:, 0], X_test[:, 1], c=y_test, cmap=plt.cm.RdBu)
    plt.title('Полиномиальная регрессия\n(Score: {:.2f})'.format(poly_score))

    plt.subplot(133)
    plt.scatter(X_test[:, 0], X_test[:, 1], c=y_test, cmap=plt.cm.RdBu)
    plt.title('Многослойный персептрон\n(Score: {:.2f})'.format(mlp_score))

    plt.tight_layout()

    plt.savefig('static/models_comparison.png')

    return render_template('index.html', result=True)

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