Árbol de Decisión modelo de Clasificación en Scikit-Learn - Titanic¶
In [3]:
from google.colab import drive
import os
drive.mount('/content/gdrive')
# Establecer ruta de acceso en dr
import os
print(os.getcwd())
os.chdir("/content/gdrive/My Drive")
Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount("/content/gdrive", force_remount=True).
/content/gdrive/My Drive
In [4]:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
titanic = pd.read_csv("Titanic.csv", sep = ",")
In [5]:
titanic
Out[5]:
| Survived | Pclass | Sex | Age | SibSp | Parch | |
|---|---|---|---|---|---|---|
| 0 | 0 | 3 | 0 | 22.0 | 1 | 0 |
| 1 | 1 | 1 | 1 | 38.0 | 1 | 0 |
| 2 | 1 | 3 | 1 | 26.0 | 0 | 0 |
| 3 | 1 | 1 | 1 | 35.0 | 1 | 0 |
| 4 | 0 | 3 | 0 | 35.0 | 0 | 0 |
| ... | ... | ... | ... | ... | ... | ... |
| 709 | 0 | 3 | 1 | 39.0 | 0 | 5 |
| 710 | 0 | 2 | 0 | 27.0 | 0 | 0 |
| 711 | 1 | 1 | 1 | 19.0 | 0 | 0 |
| 712 | 1 | 1 | 0 | 26.0 | 0 | 0 |
| 713 | 0 | 3 | 0 | 32.0 | 0 | 0 |
714 rows × 6 columns
In [6]:
#Separación en Train y Test
X = titanic.drop("Survived", axis=1)
y = titanic.Survived
In [7]:
from sklearn.model_selection import train_test_split
#Train y Test Split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30, random_state=42)
In [8]:
#Arbol de Decision
from sklearn.tree import DecisionTreeClassifier
arbol_de_decision = DecisionTreeClassifier(max_depth=4, random_state = 42)
In [9]:
#Fit
arbol_de_decision.fit(X_train,y_train) #Entrenamos el modelo
Out[9]:
DecisionTreeClassifier(max_depth=4, random_state=42)
In [10]:
#Prediccion
y_test_pred = arbol_de_decision.predict(X_test)
In [11]:
y_test_pred
Out[11]:
array([0, 1, 1, 1, 0, 0, 0, 1, 0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 1,
1, 1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0,
0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 0, 0, 1, 0, 1, 0, 0, 0,
0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0,
0, 0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 0, 1, 1, 0, 0, 1, 0, 0, 1, 0, 1,
0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0,
0, 1, 1, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1,
0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1,
0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 0, 0,
1, 0, 0, 0, 0, 1, 0, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0])
In [12]:
y_test
Out[12]:
120 0
329 1
39 1
294 1
654 0
..
534 0
393 0
382 1
223 0
140 0
Name: Survived, Length: 215, dtype: int64
A lo largo de este notebook, se solicita calcular las métricas requeridas como así también su correspondiente interpretación:
- Calcular la métrica Accuracy.
In [13]:
###Completar
from sklearn.metrics import accuracy_score
accuracy_score(y_test,y_test_pred)
Out[13]:
0.7674418604651163
- Crear la Matriz de Confusión
In [14]:
###Completar
from sklearn.metrics import confusion_matrix
confusion_matrix(y_test, y_test_pred)
Out[14]:
array([[102, 24],
[ 26, 63]])
- Calcular la métrica Precision
In [15]:
###Completar
from sklearn.metrics import precision_score
precision_score(y_test, y_test_pred)
Out[15]:
0.7241379310344828
- Calcular la métrica Recall
In [16]:
###Completar
from sklearn.metrics import recall_score
recall_score(y_test, y_test_pred)
Out[16]:
0.7078651685393258
- Calcular la métrica F1 score
In [17]:
###Completar
from sklearn.metrics import f1_score
f1_score(y_test, y_test_pred)
Out[17]:
0.7159090909090908
- classification_report
In [18]:
from sklearn.metrics import classification_report
print(classification_report(y_test,y_test_pred))
precision recall f1-score support
0 0.80 0.81 0.80 126
1 0.72 0.71 0.72 89
accuracy 0.77 215
macro avg 0.76 0.76 0.76 215
weighted avg 0.77 0.77 0.77 215
- Calcular y plotear la Curva ROC
In [19]:
###Completar
from sklearn.metrics import roc_curve, roc_auc_score
import matplotlib.pyplot as plt
y_score1 = arbol_de_decision.predict_proba(X_test)[:,1]
y_score1
Out[19]:
array([0.03797468, 1. , 0.91666667, 0.98360656, 0.03797468,
0.1474359 , 0.15789474, 0.91666667, 0.15789474, 0.66666667,
0.98360656, 0.91666667, 0.54285714, 0.03797468, 0.98360656,
0.1474359 , 0.1474359 , 0.98360656, 0.1474359 , 0.98360656,
0.16666667, 0.91666667, 0.66666667, 0.54285714, 0.1474359 ,
0.1474359 , 0.1474359 , 0.1474359 , 0.66666667, 0.03797468,
0.66666667, 0.66666667, 0.54285714, 0.03797468, 0.03797468,
0.15789474, 0.03797468, 0.21621622, 0.21621622, 0.03797468,
0.1474359 , 0.1474359 , 0.98360656, 0.03797468, 0.21621622,
0.91666667, 0.91666667, 0.66666667, 0.54285714, 0.98360656,
0.54285714, 0.91666667, 0.54285714, 0.1474359 , 0.1474359 ,
0.91666667, 0.54285714, 0. , 0.1474359 , 0.03797468,
0.98360656, 0.1474359 , 0.91666667, 0.1474359 , 0.1474359 ,
0.21621622, 0.1474359 , 0.98360656, 0.1474359 , 0.54285714,
0.1474359 , 0.98360656, 0.1474359 , 0.1474359 , 0.03797468,
0.1474359 , 0.1474359 , 0.1474359 , 0.15789474, 0.54285714,
0.16666667, 0.54285714, 0.15789474, 0.1474359 , 0.1474359 ,
0.66666667, 0.16666667, 0.03797468, 0.1474359 , 0.1474359 ,
0.21621622, 0.98360656, 0.91666667, 0.1474359 , 0.91666667,
0.66666667, 0.98360656, 0.54285714, 0.1474359 , 0.15789474,
0.91666667, 0.91666667, 0.1474359 , 0.16666667, 0.91666667,
0.03797468, 0.1474359 , 0.54285714, 0.03797468, 0.91666667,
0.1474359 , 0.1474359 , 0.1474359 , 0.21621622, 0.1474359 ,
0.15789474, 0.1474359 , 0.66666667, 0.1474359 , 0.03797468,
0.98360656, 0.1474359 , 0.1474359 , 0.1474359 , 0.98360656,
0.1474359 , 0.1474359 , 0.91666667, 0.1474359 , 0.98360656,
0.1474359 , 0.1474359 , 0.1474359 , 0.91666667, 1. ,
0.1474359 , 0.03797468, 0.54285714, 0.1474359 , 0. ,
0.03797468, 0.91666667, 0.98360656, 0.15789474, 0.1474359 ,
0.1474359 , 0.03797468, 0.1474359 , 0.54285714, 0.1474359 ,
0.03797468, 0.15789474, 0.1474359 , 0.66666667, 0.1474359 ,
0.1474359 , 0.16666667, 0.1474359 , 0.91666667, 0.21621622,
0.1474359 , 0.1474359 , 0.1474359 , 0.98360656, 0.03797468,
0.91666667, 0.98360656, 0.1474359 , 0.98360656, 0.1474359 ,
0.21621622, 0.91666667, 0.66666667, 0.91666667, 0.54285714,
0.91666667, 0.03797468, 0.1474359 , 0.03797468, 0.03797468,
0.66666667, 0.15789474, 0.98360656, 0.54285714, 0.1474359 ,
0.1474359 , 0.98360656, 0.54285714, 0.91666667, 0.03797468,
0.1474359 , 0.66666667, 0.91666667, 0.66666667, 0.66666667,
0.54285714, 0.1474359 , 0.03797468, 0.98360656, 0.15789474,
0.1474359 , 0.1474359 , 0.1474359 , 1. , 0.1474359 ,
0.1474359 , 0.66666667, 0.66666667, 0.03797468, 0.98360656,
0.1474359 , 0.21621622, 0.66666667, 0.1474359 , 0.21621622])
In [20]:
# Calculo de tasas
false_positive_rate1, true_positive_rate1, threshold1 = roc_curve(y_test, y_score1)
In [21]:
print('roc_auc_score for DecisionTree: ', roc_auc_score(y_test, y_score1))
roc_auc_score for DecisionTree: 0.8243267344390939
In [22]:
plt.subplots(1, figsize=(8,6))
plt.title('Receiver Operating Characteristic - DecisionTree')
plt.plot(false_positive_rate1, true_positive_rate1)
plt.plot([0, 1], ls="--")
plt.plot([0, 0], [1, 0] , c=".7"), plt.plot([1, 1] , c=".7")
plt.ylabel('True Positive Rate')
plt.xlabel('False Positive Rate')
plt.show()
Created in