Titanic

import pandas as pd import numpy as np import seaborn as sns import matplotlib.pyplot as plt import scipy as sc

df_train = pd.read_csv('./Titanic/train.csv') df_test = pd.read_csv('./Titanic/test.csv') pd.options.display.float_format = '{:,.2f}'.format print(df_train.shape) df_train.info()

(891, 12)
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 12 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   PassengerId  891 non-null    int64  
 1   Survived     891 non-null    int64  
 2   Pclass       891 non-null    int64  
 3   Name         891 non-null    object 
 4   Sex          891 non-null    object 
 5   Age          714 non-null    float64
 6   SibSp        891 non-null    int64  
 7   Parch        891 non-null    int64  
 8   Ticket       891 non-null    object 
 9   Fare         891 non-null    float64
 10  Cabin        204 non-null    object 
 11  Embarked     889 non-null    object 
dtypes: float64(2), int64(5), object(5)
memory usage: 83.7+ KB

df_train.head(5)

PassengerIdint64

Survivedint64

df_train.describe(include='all').fillna('-')

PassengerIdobject

-27.3%

891.018.2%

5 others54.5%

Survivedobject

-27.3%

0.027.3%

4 others45.5%

count

891.0

unique

top

freq

mean

446.0

0.3838383838383838

std

257.3538420152301

0.4865924542648575

min

1.0

0.0

25%

223.5

0.0

50%

446.0

0.0

75%

668.5

1.0

df_train.groupby(['Survived'])['Survived'].count()

sns.countplot(x=df_train['Survived'])

df_train.groupby(['Survived', 'Sex'])['Sex'].count()

sns.countplot(data=df_train, x='Sex', hue='Survived')

edad = { 'Age': [min,max, np.mean, np.median]} df_train.groupby(['Survived', 'Sex']).aggregate(edad)

Age minfloat64

Age maxfloat64

(0, 'female')

(0, 'male')

(1, 'female')

0.75

(1, 'male')

0.42

sns.boxplot(data=df_train, x='Survived', y='Age', hue='Sex')

niños = df_train[df_train['Age'] < 18.0] num_niños = len(niños['PassengerId'])/len(df_train['Age'].notnull())*100 print(f'Porcentage of kids a board {num_niños} \nHow many kids are? {len(niños)}') niños.groupby(['Survived', 'Sex']).aggregate(edad)

Porcentage of kids a board 12.682379349046016 
How many kids are? 113

Age minfloat64

Age maxfloat64

(0, 'female')

(0, 'male')

(1, 'female')

0.75

(1, 'male')

0.42

df_train.groupby('Pclass')['Pclass'].count()

ticket_class = df_train.groupby(['Survived', 'Pclass'])['Survived'].count() print(ticket_class) print(ticket_class.groupby(level=0).apply(lambda x: x/ x.sum() *100)) print(ticket_class.groupby(level=1).apply(lambda x: x/ x.sum() *100))

Survived  Pclass
0         1          80
          2          97
          3         372
1         1         136
          2          87
          3         119
Name: Survived, dtype: int64
Survived  Pclass
0         1        14.57
          2        17.67
          3        67.76
1         1        39.77
          2        25.44
          3        34.80
Name: Survived, dtype: float64
Survived  Pclass
0         1        37.04
          2        52.72
          3        75.76
1         1        62.96
          2        47.28
          3        24.24
Name: Survived, dtype: float64

sns.countplot(data=df_train, x='Survived', hue='Pclass')

df_train.groupby(['Survived', 'Pclass'])['Age'].mean() df_train.groupby(['Survived', 'Pclass'])['Age'].mean().groupby(level=1).apply(lambda x: np.mean(x))

sns.boxplot(x=df_train['Pclass'], y=df_train['Age'])

df_train.groupby(['Survived', 'Embarked'])['Survived'].count()

sns.countplot(data=df_train, x='Embarked', hue='Survived')

df_train.groupby(['Survived', 'SibSp', 'Parch'])['Survived'].count().unstack().fillna('-')

0object

-41.7%

374.08.3%

6 others50%

1object

-33.3%

13.08.3%

7 others58.3%

(0, 0)

374.0

13.0

(0, 1)

59.0

23.0

(0, 2)

12.0

1.0

(0, 3)

7.0

(0, 4)

9.0

(0, 5)

(0, 8)

(1, 0)

163.0

25.0

(1, 1)

64.0

34.0

(1, 2)

4.0

6.0

sns.barplot(data=df_train, x='SibSp', y='Parch', hue='Survived')

df_test['Family'] = np.where(((df_test['Parch'] > 0) & (df_test['SibSp'] > 0)), 1, 0) df_train['Family'] = np.where(((df_train['Parch'] > 0) & (df_train['SibSp'] > 0)), 1, 0) df_train['Family'].value_counts()

sns.countplot(data=df_train, x='Family', hue='Survived')

pd.cut(df_train['Fare'], bins= 3).value_counts()

df_test['Cost'] = pd.cut(df_test['Fare'], bins= [-1,171,342,513]) df_train['Cost'] = pd.cut(df_train['Fare'], bins= [-1,171,342,513]) df_train['Cost'].sample(5)

sns.countplot(data=df_train, x='Cost', hue='Survived')

df_train.groupby(['Survived', 'Cost'])['Survived'].count()

df_train.isnull().sum()

sns.displot(df_train['Age'].dropna(), kde=True, color='darkblue')

df_train.groupby('Pclass')['Age'].mean()

class_age = { 1:38, 2:29, 3:25 }

def fill_age(columns): age = columns[0] pclass = np.int64(columns[1]) return class_age[pclass] if pd.isnull(age) else age

df_test['Age'] = df_test[['Age', 'Pclass']].apply(fill_age, axis=1) df_train['Age'] = df_train[['Age', 'Pclass']].apply(fill_age, axis=1) df_train.isnull().sum()

df_test = df_test.drop(columns='Cabin') df_train_clean = df_train.drop(columns='Cabin') df_train_clean.sample(5)

PassengerIdint64

Survivedint64

548

549

682

683

827

828

387

388

df_train_clean.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 13 columns):
 #   Column       Non-Null Count  Dtype   
---  ------       --------------  -----   
 0   PassengerId  891 non-null    int64   
 1   Survived     891 non-null    int64   
 2   Pclass       891 non-null    int64   
 3   Name         891 non-null    object  
 4   Sex          891 non-null    object  
 5   Age          891 non-null    float64 
 6   SibSp        891 non-null    int64   
 7   Parch        891 non-null    int64   
 8   Ticket       891 non-null    object  
 9   Fare         891 non-null    float64 
 10  Embarked     889 non-null    object  
 11  Family       891 non-null    int64   
 12  Cost         891 non-null    category
dtypes: category(1), float64(2), int64(6), object(4)
memory usage: 84.8+ KB

df_test[['T_cheap', 'T_medium', 'T_expensive']] = pd.get_dummies(df_test['Cost']) df_train_clean[['T_cheap', 'T_medium', 'T_expensive']] = pd.get_dummies(df_train_clean['Cost'])

df_train_clean[['Survived', 'Pclass', 'Family', 'Cost', 'Sex']] = df_train_clean[['Survived', 'Pclass', 'Family', 'Cost', 'Sex']].astype('category') df_test[['Pclass', 'Family', 'Cost', 'Sex']] = df_test[['Pclass', 'Family', 'Cost', 'Sex']].astype('category')

df_train_clean.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 16 columns):
 #   Column       Non-Null Count  Dtype   
---  ------       --------------  -----   
 0   PassengerId  891 non-null    int64   
 1   Survived     891 non-null    category
 2   Pclass       891 non-null    category
 3   Name         891 non-null    object  
 4   Sex          891 non-null    category
 5   Age          891 non-null    float64 
 6   SibSp        891 non-null    int64   
 7   Parch        891 non-null    int64   
 8   Ticket       891 non-null    object  
 9   Fare         891 non-null    float64 
 10  Embarked     889 non-null    object  
 11  Family       891 non-null    category
 12  Cost         891 non-null    category
 13  T_cheap      891 non-null    uint8   
 14  T_medium     891 non-null    uint8   
 15  T_expensive  891 non-null    uint8   
dtypes: category(5), float64(2), int64(3), object(3), uint8(3)
memory usage: 63.5+ KB

print(pd.get_dummies(df_train_clean['Sex'])) df_test[['Female', 'Male']] = pd.get_dummies(df_test['Sex']) df_train_clean[['Female', 'Male']] = pd.get_dummies(df_train_clean['Sex']) df_train_clean[['Female', 'Male']]

     female  male
0         0     1
1         1     0
2         1     0
3         1     0
4         0     1
..      ...   ...
886       0     1
887       1     0
888       1     0
889       0     1
890       0     1

[891 rows x 2 columns]

Femaleuint8

0 - 1

Maleuint8

0 - 1

from sklearn import metrics from sklearn.linear_model import LogisticRegression from sklearn.tree import DecisionTreeClassifier

feature_columns = ['Age', 'Pclass', 'Female', 'Male', 'Family', 'T_cheap', 'T_medium', 'T_expensive'] x = df_train_clean[feature_columns] y = df_train_clean['Survived'] # Modelo de regresion logistica logreg = LogisticRegression() logreg.fit(x, y) # Modelo de arboles de decisión tree_one = DecisionTreeClassifier() tree_one.fit(x, y)

# labels = ['True Neg', 'False Pos', 'False Neg', 'True Pos'] # labels = np.asarray(labels).reshape(2,2) # Modelo de regresion logistica y_predict = logreg.predict(x) cf_matrix = metrics.confusion_matrix(y, y_predict) print(sns.heatmap(cf_matrix, annot=True, fmt='.2f', cmap='Blues', cbar=False)) print(metrics.accuracy_score(y, y_predict))

AxesSubplot(0.125,0.125;0.775x0.755)
0.8047138047138047

# Modelo de arbol de decisición y_tree_predict = tree_one.predict(x) cf_matrix_tree = metrics.confusion_matrix(y, y_tree_predict) print(sns.heatmap(cf_matrix_tree, annot=True, fmt='.2f', cmap='Greens', cbar=False)) print(metrics.accuracy_score(y, y_tree_predict))

AxesSubplot(0.125,0.125;0.775x0.755)
0.8967452300785634

def conf_mat_acc(modelo): disp = metrics.plot_confusion_matrix(modelo, x, y, cmap=plt.cm.Blues, values_format="d") true_pred = disp.confusion_matrix[0,0]+disp.confusion_matrix[1,1] total_data = np.sum(disp.confusion_matrix) accuracy = true_pred/total_data print('accuracy: ', np.round(accuracy, 3)) plt.show()

conf_mat_acc(logreg)

/shared-libs/python3.9/py/lib/python3.9/site-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.
  warnings.warn(msg, category=FutureWarning)
accuracy:  0.805

# Modelo de regresión logistica x_test = df_test[['Age', 'Pclass', 'Female', 'Male', 'Family', 'T_cheap', 'T_medium', 'T_expensive']] x_test.to_csv('./Titanic/pruebaUnitaria.csv') df_test['Survived'] = tree_one.predict(x_test) data_send = df_test[['PassengerId', 'Survived']] data_send.to_csv('answer.csv', index=False)

# Prueba unitaria df_prueba = pd.read_csv('./Titanic/pruebaUnitaria.csv') x_test = df_prueba[['Age', 'Pclass', 'Female', 'Male', 'Family', 'T_cheap', 'T_medium', 'T_expensive']] df_prueba['Survived'] = tree_one.predict(x_test) data_send = df_prueba['Survived'] data_send.to_csv('./Titanic/pruebaUnitariaRespuesta.csv', index=False)

%pylab inline import matplotlib.pyplot as plt import matplotlib.image as mpimg

Populating the interactive namespace from numpy and matplotlib

# Visualizacion de superviviente def visualizacion(respuesta): if respuesta == 1: img = mpimg.imread('./Image/success.png') else: img = mpimg.imread('./Image/fail.png') plt.imshow(img) plt.title("¿Sobreviviste?", {'fontsize':24}) plt.axis(False) plt.show()

resultado = pd.read_csv('./Titanic/pruebaUnitariaRespuesta.csv') visualizacion(resultado['Survived'][0])