Kaggle入门竞赛-Titanic成员幸存

# This Python 3 environment comes with many helpful analytics libraries installed
# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python
# For example, here's several helpful packages to load

import numpy  # linear algebra
import pandas # data processing, CSV file I/O (e.g. pd.read_csv)

# Input data files are available in the read-only "../input/" directory
# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory

import os
for dirname, _, filenames in os.walk('/kaggle/input'):
    for filename in filenames:
        print(os.path.join(dirname, filename))

# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using "Save & Run All" 
# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session

train_data = pandas.read_csv('/kaggle/input/titanic/train.csv')
train_data.head()

train_data.info()

age_median = train_data["Age"].median()
train_data["Age"].fillna(age_median, inplace = True)
train_data.info()

#训练集预测器和标签分离
train_survived = train_data["Survived"]
features = ["Pclass", "Sex", "SibSp", "Parch", "Age"]
train_features = pandas.get_dummies(train_data[features])

#学习随机森林模型
from sklearn.ensemble import RandomForestClassifier

forest_model = RandomForestClassifier(random_state = 24)
forest_model.fit(train_features, train_survived)

#模型随机搜索参数
from sklearn.model_selection import RandomizedSearchCV

#随机搜索最优参数
param_rand = [
    {"n_estimators":range(10,1000,20), "max_depth":range(1,6,1), "bootstrap": [True, False]}
] 

random_search = RandomizedSearchCV(forest_model, param_distributions = param_rand, n_iter = 20, cv = 3)
random_search.fit(train_features, train_survived)

forest_final_model = random_search.best_estimator_

#执行K折交叉验证
from sklearn.model_selection import cross_val_predict
#ROC曲线
train_probas_forest = cross_val_predict(forest_final_model, train_features, train_survived,cv = 3, method = "predict_proba")
train_scores_forest = train_probas_forest[:,1]

from sklearn.metrics import roc_curve
fpr_forest, tpr_forest, threholds_forest = roc_curve(train_survived, train_scores_forest)

import matplotlib.pyplot

def plot_roc_curve(fpr, tpr, label = None):
    matplotlib.pyplot.plot(fpr, tpr, linewidth = 2, label = label)
    matplotlib.pyplot.plot([0, 1], [0, 1], "k--")
    [...]

plot_roc_curve(fpr_forest, tpr_forest)
matplotlib.pyplot.show()

#ROC分数
from sklearn.metrics import roc_auc_score
roc_auc_score(train_survived, train_scores_forest)

from sklearn.linear_model import SGDClassifier
sgd_model = SGDClassifier(random_state = 24)
sgd_model.fit(train_features, train_survived)

train_scores_sgd = cross_val_predict(sgd_model, train_features, train_survived, cv = 3, method = "decision_function")
fpr_sgd, tpr_sgd, threholds_sgd = roc_curve(train_survived, train_scores_sgd)
plot_roc_curve(fpr_sgd, tpr_sgd)
matplotlib.pyplot.show()

roc_auc_score(train_survived, train_scores_sgd)

test_data = pandas.read_csv('/kaggle/input/titanic/test.csv')
test_data.head()

test_data.info()

test_data["Age"].fillna(age_median, inplace = True)
test_features = pandas.get_dummies(test_data[features])
Last_predic = forest_final_model.predict(test_features)

csv_out = pandas.DataFrame({'PassengerId': test_data.PassengerId, 'Survived': Last_predic})
csv_out.to_csv('my_submission.csv', index = False)
print("Your submission was successfully saved!")