from flask import Flask, render_template, request
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.neural_network import MLPRegressor
from sklearn.pipeline import Pipeline
import joblib

app = Flask(__name__)

# Загрузка данных
data_bgg = pd.read_csv("bgg_dataset.csv", delimiter=";")

# Выбор нужных столбцов
selected_columns_bgg = ['Year Published', 'Users Rated', 'Rating Average', 'BGG Rank', 'Owned Users', 'Complexity Average']
features = data_bgg[selected_columns_bgg]

# Замена запятых на точки в столбцах 'Rating Average' и 'Complexity Average'
features['Rating Average'] = features['Rating Average'].str.replace(',', '.').astype(float)
features['Complexity Average'] = features['Complexity Average'].str.replace(',', '.').astype(float)

# Замена пропущенных значений средними значениями по столбцам
features = features.fillna(features.mean())

# Разделение данных
X_bgg = features.drop('Rating Average', axis=1)
y_bgg = features['Rating Average']
X_train_bgg, X_test_bgg, y_train_bgg, y_test_bgg = train_test_split(X_bgg, y_bgg, test_size=0.2, random_state=42)

# Масштабирование данных
scaler = StandardScaler()
X_train_bgg_scaled = scaler.fit_transform(X_train_bgg)
X_test_bgg_scaled = scaler.transform(X_test_bgg)

# Обучение модели нейронной сети
mlp_regressor_model = Pipeline([
    ('scaler', StandardScaler()),
    ('mlp_regressor', MLPRegressor(hidden_layer_sizes=(100, 50), max_iter=2000, random_state=42))
])
mlp_regressor_model.fit(X_train_bgg_scaled, y_train_bgg)

# Сохранение модели
joblib.dump(mlp_regressor_model, 'mlp_regressor_model.joblib')

# Загрузка модели
mlp_regressor_model = joblib.load('mlp_regressor_model.joblib')

# Обновление маршрута для предсказания
@app.route('/')
def index():
    return render_template('index.html')

@app.route('/predict', methods=['POST'])
def predict():
    if request.method == 'POST':
        # Получение данных из формы
        input_data_bgg = {
            'Year Published': int(request.form['Year Published']),
            'Users Rated': int(request.form['Users Rated']),
            'BGG Rank': int(request.form['BGG Rank']),
            'Owned Users': int(request.form['Owned Users']),
            'Complexity Average': float(request.form['Complexity Average'])
        }

        # Преобразование данных в DataFrame
        input_df_bgg = pd.DataFrame([input_data_bgg])

        # Масштабирование входных данных
        input_data_scaled = scaler.transform(input_df_bgg)

        # Предсказание
        prediction_bgg = mlp_regressor_model.predict(input_data_scaled)[0]

        return render_template('index.html', prediction_bgg=prediction_bgg)

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