# -*- coding: utf-8 -*-
"""Proyecto Final Juan Mosquera.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1x--JFmF7BoNV0jWxGBv-2a4gnH67eJ1J
"""

# Cargar librerias
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

# Detección de valores vacios
import missingno as msno # !pip install missingno

# Libreria para valores aleatorios
import random

# Librerias de Machine Learning
from sklearn.model_selection import train_test_split
from imblearn.over_sampling import SMOTE
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score, confusion_matrix, classification_report
import xgboost as xgb
from sklearn.ensemble import RandomForestClassifier

pwd

# Cargar Datos
df = pd.read_csv('Dataset_Infartos.csv',sep=';')
df.head()

"""En este caso podemos ver las columnas que tiene nuestro dataset siempre mostrando las primeras 5 columnas, podemos ver desde el principio que la mayoria de las columnas tienen valores correspondientes aunque en la columna de Flag_fumador existe valores vacios o NaN, es necesario hacer una limpieza ya sea viendo la cantidad de NaN que existen."""

# conociendo el dimensionamiento
df.shape

df.info()

"""Aca vemos la cantidad de filas que tiene cada columna y vemos que la mayoria cuenta con 43400 datos, sin embargo, para la columnna IMC vemos que tiene 1462 datos nulls y Flag_fumador tiene 13292 datos nulls, para cada uno equivale un 3,4% y 30,6%"""

# valores vacios
df.isna().sum() / 43400

"""Desde aca podemos hacer nuestro primer analisis con respecto a estas dos columnas, decimos que "Si el valor del porcentaje es menor al 10% o 15% máximo, se puede hacer una imputación para los valores por ejemplo para IMC y si es superior se realiza la eliminación de la variable por ejemplo para Flag_fumador"""

# Ver la correlacion o relación de los vacios
msno.matrix(df, figsize=(20, 4))

df['Ataque_cardiaco'].value_counts(1)

"""Vemos algo muy importante y es que para nuestra variable target la proporcion esta desbalanceada, posiblemente sera necesario usar una tecnica de balaceo con los datos"""

df.describe()

"""Tenemos que hacer la comparación del promedio y de la desviación estándar, verificar que el promedio sea mayor que la desviación estándar, en el caso de Flag_Hipertension-Flag_problem_cardiaco vemos que tienen 0 y 1"""

# Graficar histograma de las variables Edad, glucosa y IMC
df[['Edad','Promedio_nivel_glucosa','IMC']].hist(bins=20,figsize=(15,8))
plt.show()

axes = df[['Edad', 'Promedio_nivel_glucosa', 'IMC']].hist(bins=20, figsize=(15, 8))

# Agregar una línea vertical en la media de cada variable
for ax, col in zip(axes.flatten(), ['Edad', 'Promedio_nivel_glucosa', 'IMC']):
    mean_value = df[col].mean()
    ax.axvline(mean_value, color='red', linestyle='dashed', linewidth=2, label=f'Media: {mean_value:.2f}')
    ax.legend()  # Mostrar la leyenda con la media

plt.show()

"""Al saber cada una de las medias podemos ver por donde se estan agrupando mas los valores correspondientes, como vimos en clase el nivel de glucosa es muy importante para identificar si es diabetico, Menos de 100 mg/dL se considera normal. Entre 100 y 125 o más en dos pruebas distintas se diagnostica como diabetes. Por otro lado, la edad esta entre los 40-60 años donde se almacena la mayor cantidad de datos, al buscar el Indice de masa corporal se puede ver que se usa para identificar las categorías de peso que pueden llevar a problemas de salud, si se tuviera una variables extra como seria la estatura se podria calcular cual seria el IMC adecuado para cada una de las personas pero vemos que los datos estan entre el 20-35."""

# Analisis multivariable
sns.pairplot(df[['Edad','Promedio_nivel_glucosa','IMC','Ataque_cardiaco']], hue='Ataque_cardiaco')
plt.show()

"""En la Edad como en el IMC vemos que para los ataques cardiacos estan dirigidos para un lado en especifico, suele pasar para las personas que son adultas y para las personas que tienen un nivel 22 y 35, si se comparan las demas variables la unica donde podemos ver que tiene informacion significante es la relacion entre Edad y Nivel de glucosa, donde se ve muy claro que los ataques cardiacos afectan mas a las personas adultas y no importa que nivel de glucosa tienen. Tambien se puede ver el IMC vs Nivel de glucosa, vemos que los puntos naranjas estan distribuidos por todo el diagrama pero se puede ver una linea recta donde los ataques cardiacos son frecuentas si el IMC es mayor a 25."""

# Calcular correlaciones
df[['Edad','Promedio_nivel_glucosa','IMC','Ataque_cardiaco']].corr()

"""Aca podemos revisar la correlacion para el ataque cardiaco que es nuestro Target, siempre debemos tener presente que entre mas cercano sea al 1 mayor es su correlacion, por lo tanto, la variable mas importante es Edad-Nivel Glucosa-IMC."""

df[['Edad','Promedio_nivel_glucosa','IMC','Ataque_cardiaco','Flag_hipertension','Flag_problem_cardiaco']].corr()

"""Aca podemos ver lo mismo de arriba solo que con mas variables y si queremos escoger solo tres variaboles importamtes serian las que tenemos en el anterior codigo."""

df.select_dtypes(include='object').head(2)

"""Para tomar una decision si una variable es relevante o significativa vemos la meida de cada una de las variables con respecto a los ataques cardiacos, para grandes volumenes de datos si el porcentaje que subio es mayor al 5% podemos decir que es relevante la variable, por lo contrario si el volumen de datos es mucho mas corto buscamos que sea superior a ese 5%, que sea un 10%."""

# Conocer la distribución de mi variable
df['Genero'].value_counts(1)

# Ataque cardiaco por Genero
df[['Genero','Ataque_cardiaco']].groupby(by='Genero').agg(['count','sum','mean'])

"""Teniendo presente la proporcion de ataques cardiacos que es de 1,8%, comparamos el genero y vemos que la proporcion que subio solo fue el 0,1% por lo que el genero no es relevante para decidir si tiene ataques cardiacos o no."""

# Conocer la distribución de mi variable
df['Estados_civil'].value_counts(1)

# Ataque cardiaco por Estados_civil
df[['Estados_civil','Ataque_cardiaco']].groupby(by='Estados_civil').agg(['count','sum','mean'])

"""Teniendo presente la proporcion de ataques cardiacos que es de 1,8%, comparamos el estados civil y vemos que la proporcion que subio fue el 0,7122% que equivale al 39,5% por lo que el estados civil si puede ser una variable relevante para decidir si tiene ataques cardiacos o no. En conclusion las personas casadas son las que tienen mas ataques cardiacos."""

# Conocer la distribución de mi variable
df['Tipo_trabajo'].value_counts(1)

# Ataque cardiaco por Nunca_trabajo
df[['Tipo_trabajo','Ataque_cardiaco']].groupby(by='Tipo_trabajo').agg(['count','sum','mean'])

"""Si seguimos el mismo orden el tipo de trabajo es una variable significativa ya que podemos ver que para los emprendedores su promedio es alto superando la proporcion del ataque vcardiaco, los demas trabajos se mantiene igual excepto Nunca tabaja y Cuidar ninos."""

# Conocer la distribución de mi variable
df['Zona_residencia'].value_counts(1)

# Ataque cardiaco por Nunca_trabajo
df[['Zona_residencia','Ataque_cardiaco']].groupby(by='Zona_residencia').agg(['count','sum','mean'])

"""Para la variable zona de residencia vemos que el promedio no cambia, por lo tanto, es una variable que no influye mucho a la hora de ver si causa ataques cardiacos."""

# Conocer la distribución de mi variable
df['Flag_fumador'].value_counts(1)

# Conocer la distribución de mi variable
df['Flag_fumador'].value_counts(1,dropna=False)

"""Vemos que esta variable Flag fumador contiene muchos datos vacios por lo tanto no sabemos como se podra estar comportando esta variable"""

# Ataque cardiaco por flag_fumador
df[['Flag_fumador','Ataque_cardiaco']].groupby(by='Flag_fumador').agg(['count','sum','mean'])

"""Si queremos usar la variable Flag fumador lo que se podria hacer es crear una columna aparte donde nos coloquen 1 y 0, 1 se usa por si se encuentra un dato en la variable flag fumador y 0 por si se encuentra nulo.

Como se indico en la clase las variables importantes serian,
Variables Numericas:

Edad

*   Edad
*   Flag_problem_cardiaco
*   Promedio_nivel_glucosa
*   Flag_hipertension
*   IMC

Variables Cualitativas importantes:

*   Estados_Civil
*   Tipo_trabajo
"""

df['IMC'].describe()

"""Como lo pudimos ver desde un comienzo, la variable IMC cuenta con valores faltantes 1462 datos nulls que equivalen a 3,4%, po lo tanto, podemos realizar el metodo mas eficiente de imputacion que es el de intervalos de confianza, para lograr completar los valores nulos."""

# Promedio y desviacion del IMC
IMC_avg = df['IMC'].mean()
IMC_std = df['IMC'].std()
IMC_avg, IMC_std

# Calculo de valore svacios para IMC
IMC_nullnum = df['IMC'].isna().sum()
IMC_nullnum

# Generar la lista de valores a reemplazar
#np.random.randint(IMC_avg - IMC_std, IMC_avg +  IMC_std,size = IMC_nullnum)
random.seed(123)
IMC_lista_vacios = np.random.uniform(IMC_avg - IMC_std, IMC_avg +  IMC_std,size = IMC_nullnum)

# Filtrar valores vacios en df['IMC']
df.loc[np.isnan(df['IMC']), 'IMC']

# Reemplazar los valores vacios por la lista de arriba
df.loc[np.isnan(df['IMC']), 'IMC'] = IMC_lista_vacios

df['IMC'].describe()

"""Ya podemos ver que se completaron nuestro datos faltantes"""

# Transformar variable: Edad
df['Edad_Encoded'] = 0
df.loc[df['Edad']<5, 'Edad_Encoded'] = 1 # Primera Infancia
df.loc[(df['Edad']>=5) & (df['Edad']<=11), 'Edad_Encoded'] = 2 # Infancia
df.loc[(df['Edad']>=12) & (df['Edad']<=18), 'Edad_Encoded'] = 3 #Adolescencia
df.loc[(df['Edad']>=14) & (df['Edad']<=26), 'Edad_Encoded'] = 4 #Joven
df.loc[(df['Edad']>=27) & (df['Edad']<=59), 'Edad_Encoded'] = 5 #Adultez
df.loc[df['Edad']>=60, 'Edad_Encoded'] = 6 #Vejez

df['Edad_Encoded'].value_counts()

"""Es interesante en algun otro ejercicio aplicar la siguiente clasificacion de generaciones ![image.png](data:image/png;base64,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)"""

# Transformar variable de Promedio Nivel Gluca
df['Gluocosa_Encoded'] = 0
df.loc[df['Promedio_nivel_glucosa']<100, 'Gluocosa_Encoded'] = 1 # Normal
df.loc[(df['Promedio_nivel_glucosa']>=100) & (df['Promedio_nivel_glucosa']<=125), 'Gluocosa_Encoded'] = 2 # Prediabetes
df.loc[df['Promedio_nivel_glucosa']>=126, 'Gluocosa_Encoded'] = 3 # Diabetes

df['Gluocosa_Encoded'].value_counts()

# Transformar variable de IMC
df['IMC_Encoded']  = 0
df.loc[df['IMC']<18.5, 'IMC_Encoded'] = 1 # Bajo Peso
df.loc[(df['IMC']>=18.5) & (df['IMC']<=24.9), 'IMC_Encoded'] = 2 # Peso Saludable
df.loc[(df['IMC']>=25) & (df['IMC']<=29.9), 'IMC_Encoded'] = 3 # Sobrepeso
df.loc[df['IMC']>=30, 'IMC_Encoded'] = 4 # Obesidad

df['IMC_Encoded'].value_counts()

# Transformar mis variables Cualitativas: Estados_Civil
df['Estado_Encoded'] = df['Estados_civil'].map({'Si':1,'No':2,'':2}).astype(int)

df['Estado_Encoded'].value_counts()

"""indicamos si en la variable estado civil dice si entonces pongale 1, si dice no pongale 2 y si esta vacia pongale 2"""

# Transformar mis variables Cualitativas: Tipo Trabajo
df['TipoTrabajo_Encoded'] = df['Tipo_trabajo'].map({'Emprendedor':1,
                                                    'Empresa_privada':2,
                                                    'En_gobierno':3,
                                                    'Nunca_trabajo':4,
                                                    'cuidar_ninos':4,
                                                    '':4}).astype(int)

df['TipoTrabajo_Encoded'].value_counts()

df.head()

"""Ya encontramos nuestras variables agregadas en la tabla principal para hacer el modelo correspondiente"""

# Filtrar valores transformados + variables numericas
df_encoded = df[['Ataque_cardiaco','Flag_hipertension','Flag_problem_cardiaco',
                 'Edad_Encoded','Gluocosa_Encoded','IMC_Encoded','Estado_Encoded','TipoTrabajo_Encoded']]

"""seleccionamos las variables que usaremos y las guardamos en una nueva tabla"""

df_encoded.head()

df_encoded['Ataque_cardiaco'].value_counts(1)

"""Tenemos nuestros datos desbalanceados ya sea nuestro numero 0 el numero mayoritario y el 1 minoritario"""

X = df_encoded.drop('Ataque_cardiaco',axis=1)
y = df_encoded['Ataque_cardiaco']

X.head()

y.head()

# Calcular mi dataset de entrenamiento y test
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=123)

len(X_train), len(X_test), len(y_train), len(y_test)

df_train = pd.concat([X_train,y_train],axis=1)
df_train.head()

count_class_0, count_class_1 = df_train['Ataque_cardiaco'].value_counts()
print(count_class_0, count_class_1)

df_train_class_0 = df_train[df_train['Ataque_cardiaco'] == 0]
df_train_class_1 = df_train[df_train['Ataque_cardiaco'] == 1]

count_class_0, count_class_1, df_train_class_0.shape, df_train_class_1.shape

"""Contamos con tres metodos de balanceo

1.   UnderSampling
2.   OverSampling
3.   SMOOTE

Veremos como se comporta cada uno de los metodos en nuestro dataset

Usaremos primero el Metodo UnderSampling
"""

# Obtener una muestra de cantidad de la categoria minoritaria del dataset de la clase mayoritaria
df_train_class_0_under = df_train_class_0.sample(count_class_1,random_state = 123)

# Concatenar el datset obtenido con el dataset de la clase minoritaria
df_train_u = pd.concat([df_train_class_0_under, df_train_class_1], axis=0)

# Revisión de la distribución de la variable ataque_cardiaco
df_train_u['Ataque_cardiaco'].value_counts(1)

df_train_u['Ataque_cardiaco'].value_counts()

"""De esta misma forma funciona OverSampling solo que ahora se hace a la inversa"""

# Obtener una muestra de cantidad de la categoria mayoritaria del dataset de la clase minoriatia
df_train_class_1_over = df_train_class_1.sample(count_class_0,random_state = 123,replace = True)

# Concatenar el datset obtenido con el dataset de la clase minoritaria
df_train_o = pd.concat([df_train_class_1_over, df_train_class_0], axis=0)

# Revisión de la distribución de la variable ataque_cardiaco
df_train_o['Ataque_cardiaco'].value_counts(1)

df_train_o['Ataque_cardiaco'].value_counts()

"""Por ultimo veremos el metodo SMOOTE"""

# Ejemplo Smoote
df_ejemplo = pd.concat([df_train[df_train['Ataque_cardiaco']==0].head(5),df_train[df_train['Ataque_cardiaco']==1].head(2)],axis = 0)
df_ejemplo

# Defino mis X e Y
aux_1 = df_ejemplo.drop('Ataque_cardiaco',axis=1)
aux_2 = df_ejemplo['Ataque_cardiaco']

# Entrenar el algoritmos
smote = SMOTE(k_neighbors=1,random_state=123,sampling_strategy = 'minority')
X_smote, y_smote = smote.fit_resample(aux_1, aux_2)

# Resultados
pd.concat( [X_smote,y_smote],axis=1)

# Ejecutar SMOOTE a todos los datos
smote = SMOTE(k_neighbors=3,random_state=123,sampling_strategy = 'minority')
X_smote, y_smote = smote.fit_resample(X_train, y_train)
df_train_smote = pd.concat([X_smote,y_smote],axis=1)

df_train_smote['Ataque_cardiaco'].value_counts(1)

df_train_smote['Ataque_cardiaco'].value_counts(0)

"""Ya contamos con los tres modelos ahora lo que necesitamos es mirar que modelo tomar, cual sera el mejor de esos tres"""

df_train_u.shape, df_train_o.shape, df_train_smote.shape

# Ejecutar un modelo sencillo (ARBOL DE DECISION)
# DATA SIN BALANCEO

# 1) llamar al modelo
model = DecisionTreeClassifier(random_state=123)

# 2) Entrenar el modelo
model.fit(X_train,y_train)

# 3) Predicción
y_pred = model.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred))
print('Reporte de classification')
print(classification_report(y_test,y_pred,digits = 3))

"""Algo pasa con la precision que es 0 para nuestra categoria 1, en la matriz vemos que no reconoce ningun dato de los 4 que aparecen, entonces algo anda mal.Esto es para nuestros datos originales."""

# DATA BALANCEADOS (undersampling)

# 1) llamar al modelo
model = DecisionTreeClassifier(random_state=123)

# 1.5) Obtener X_train e y_train
X_train_u = df_train_u.drop('Ataque_cardiaco',axis=1)
y_train_u = df_train_u['Ataque_cardiaco']

# 2) Entrenar el modelo
model.fit(X_train_u,y_train_u)

# 3) Predicción
y_pred = model.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred))
print('Reporte de classification')
print(classification_report(y_test,y_pred,digits = 3))

# DATA BALANCEADOS (OverSampling)

# 1) llamar al modelo
model = DecisionTreeClassifier(random_state=123)

# 1.5) Obtener X_train e y_train
X_train_o = df_train_o.drop('Ataque_cardiaco',axis=1)
y_train_o = df_train_o['Ataque_cardiaco']

# 2) Entrenar el modelo
model.fit(X_train_o,y_train_o)

# 3) Predicción
y_pred = model.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred))
print('Reporte de classification')
print(classification_report(y_test,y_pred,digits = 3))

# DATA BALANCEADOS (SMOOTE)

# 1) llamar al modelo
model = DecisionTreeClassifier(random_state=123)

# 1.5) Obtener X_train e y_train
X_train_s = df_train_smote.drop('Ataque_cardiaco',axis=1)
y_train_s = df_train_smote['Ataque_cardiaco']

# 2) Entrenar el modelo
model.fit(X_train_s,y_train_s)

# 3) Predicción
y_pred = model.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred))
print('Reporte de classification')
print(classification_report(y_test,y_pred,digits = 3))

# Elección del Modelo
Tabla = pd.DataFrame({'Balanceo':['Original','Under','Over','Smote'],
                      'Accuracy': [98.09,74.08,73.88,73.26],
                      'Precision':[0,0.052,0.043,0.044],
                      'Recall':[0,0.741,0.617,0.636],
                      'F1':[0,0.096,0.081,0.082]})
Tabla

"""Es muy interesante analisar estos datos ya que para algunos casos es mas importante el numero de Precision y en otros el de Recall, para este ejercicio usaremos el de Recall ya que como se habla que para la agencia de seguros mandar mucha gente con precision a ataques cardiacos pues seria un gasto economico por lo tanto es mejor el Recall en este caso. El mejor de todos es el undersampling con un Recall de 0.741, en otros casos como si fueran ventas de casas o llamadas si podriamos usar Precision y el que se eligiria seria el mismo undersampling."""

# Obtener X_train e y_train
X_train_u = df_train_u.drop('Ataque_cardiaco',axis=1)
y_train_u = df_train_u['Ataque_cardiaco']

"""comenzamos con nuestros arboles de decision usando el metodo: Xgboost (Sin Tunear)"""

# 1) llamar al modelo
model1 = xgb.XGBClassifier(random_state=123)

# 2) Entrenar el modelo
model1.fit(X_train_u,y_train_u)

# 3) Predicción
y_pred1 = model1.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred1))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred1))
print('Reporte de classification')
print(classification_report(y_test,y_pred1,digits = 3))

"""Metodo Xgboost (Tuneado)"""

# 1) llamar al modelo
model2 = xgb.XGBClassifier(random_state=123,n_estimators = 10, max_depth = 4, learning_rate=0.1,subsample = 0.5)

# 2) Entrenar el modelo
model2.fit(X_train_u,y_train_u)

# 3) Predicción
y_pred2 = model2.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred2))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred2))
print('Reporte de classification')
print(classification_report(y_test,y_pred2,digits = 3))

"""Metodo Random Forest (Sin Tunear)"""

# 1) llamar al modelo
model3 = RandomForestClassifier(random_state=123)

# 2) Entrenar el modelo
model3.fit(X_train_u,y_train_u)

# 3) Predicción
y_pred3 = model3.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred3))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred3))
print('Reporte de classification')
print(classification_report(y_test,y_pred3,digits = 3))

"""Metodo Random Foreste (Tuneado)"""

# 1) llamar al modelo
model4 = RandomForestClassifier(random_state=123,n_estimators = 10 , max_depth = 4)

# 2) Entrenar el modelo
model4.fit(X_train_u,y_train_u)

# 3) Predicción
y_pred4 = model4.predict(X_test)

# 4) Evaluación
print('acurracy:')
print(accuracy_score(y_test,y_pred4))
print('Reporte de confusion_matrix')
print(confusion_matrix(y_test,y_pred4))
print('Reporte de classification')
print(classification_report(y_test,y_pred4,digits = 3))

# Elección del Modelo
Tabla1 = pd.DataFrame({'Modelos':['Xgboost (Sin Tuneo)','Xgboost (Con Tuneo)','RandomForest (Sin Tuneo)','RandomForest (Con Tuneo)','Arbol Decision'],
                      'Accuracy': [71.01, 72.99, 71.26,67.35,74.08],
                      'Precision':[0.047,0.052,0.048,0.045,0.052],
                      'Recall':[0.759,0.778,0.772,0.821,0.741],
                      'F1':[0.089,0.097,0.091,0.086,0.096]})
Tabla1

"""Para escoger que modelo es el mejor podemos revisar el Recall de nuevo y vemos que RandomForest con tuneo tiene el valor mas grande de todos, esa es la mejor opcion para nuestro modelo, en caso que no se realizara el metodo RandomForest (Con Tuneo) como en Recall vemos numero muy parecidos tendriamos que ver otras columnas como la de Accuracy y la que mejor tiene el Accuracy es Xgboost (Con Tuneo)

Por otro lado, si llegamos a tener mayor volumen de datos posiblemente el Xgboost funcione mucho mejor, para los RandomForest una cantidad no tan grande suele funcionar mejor

Prediccion con nuevos datos
"""

# Si quiero trabajar en otro computador lo que se hace es
import pickle

# Exportar modelo que nos sirve y sale en nuestras carpetas
pickle.dump(model4,open('model4.pkl','wb'))

# Si necesito cargar un modelo entonces lo que se hace es
model4.pkl = pickle.load(open('model4.pkl','rb'))

df_encoded.head(2)

# Definimos un conjunto de datos ficticios
datos_dummy = pd.DataFrame({'Flag_hipertension':[1],
                            'Flag_problem_cardiaco' : [1],
                            'Edad_Encoded':[4],
                            'Gluocosa_Encoded':[2],
                            'IMC_Encoded':[3],
                            'Estado_Encoded':[0],
                            'TipoTrabajo_Encoded':[2]})

datos_dummy

model4.predict(datos_dummy)

model4.pkl.predict(datos_dummy)

"""Entonces con los datos que coloque en la linea 112 me estan indicando que el numero 1, esto quiere decir que puede tener ataque cardiaco, entonces el seguro no le prestara su servicio o posiblemente le podria cobrar mas al cliente reduciendo la perdida de dinero, tambien lo que se puede hacer es ofrecer pólizas con primas ajustadas al riesgo, sugerir programas de salud preventiva con incentivos (descuentos por mejoras en el estilo de vida) o crear programas de detección temprana en colaboración con instituciones médicas."""