import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.compose import ColumnTransformer, TransformedTargetRegressor
from sklearn.preprocessing import StandardScaler, OrdinalEncoder, OneHotEncoder, PowerTransformer
from sklearn.pipeline import Pipeline
from sklearn.linear_model import LinearRegression 
from sklearn.ensemble import RandomForestRegressor
import sys
import os

df_origin=pd.read_csv("insurance.csv")

X, y = df.drop(['charges'], axis=1), df['charges']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=42)

# Definir las columnas por tipo
numeric_features = ['bmi', 'age', 'children']
ordinal_features = ['smoker']
categorical_features = ['sex', 'region']

# Crear el preprocessor para las features
preprocessor = ColumnTransformer([
    ('numeric', 'passthrough', numeric_features),
    ('ordinal', OrdinalEncoder(categories=[['no', 'yes']]), ordinal_features),
    ('categorical', OneHotEncoder(drop='first'), categorical_features)
])

# Crear el modelo base RandomForest
model = RandomForestRegressor(random_state=42)

# Pipeline completo con preprocessor + modelo
ml_pipeline = Pipeline([
    ('preprocessor', preprocessor),
    ('model', model)
])

# Añadir Box-Cox al target con TransformedTargetRegressor
final_model = TransformedTargetRegressor(
    regressor=ml_pipeline,
    transformer=PowerTransformer(method='box-cox')
)

param_distributions = {
    'regressor__model__n_estimators': randint(20, 200),
    'regressor__model__max_depth': [None] + list(range(5, 30)),
    'regressor__model__min_samples_split': randint(2, 20),
    'regressor__model__min_samples_leaf': randint(1, 10),
    'regressor__model__max_features': ['sqrt', 'log2', None],
    'regressor__model__bootstrap': [True, False]
}

random_search = RandomizedSearchCV(
    estimator=final_model,
    param_distributions=param_distributions,
    n_iter=50,  # Número de combinaciones a probar
    cv=5,
    scoring='neg_mean_squared_error',
    n_jobs=-1,
    verbose=1,
    random_state=42
)


random_search.fit(X_train, y_train)

# Mejor modelo encontrado
best_model = random_search.best_estimator_
print(f"\nMejores parámetros: {random_search.best_params_}")
print(f"Mejor score CV: {-random_search.best_score_:.2f}")

print("Mejores parámetros encontrados:")
for param, value in random_search.best_params_.items():
    print(f"{param}: {value}")

y_pred = best_model.predict(X_test)

mse = mean_squared_error(y_test, y_pred)
rmse = np.sqrt(mse)
r2 = r2_score(y_test, y_pred)
mae = mean_absolute_error(y_test, y_pred)

print(f"\nResultados en test set:")
print(f"MSE: {mse:.2f}")
print(f"RMSE: {rmse:.2f}")
print(f"R²: {r2:.4f}")
print(f"MAE: {mae:.2f}")