Obsidian-Knowledge-Base/X-Knowledge-Base/raw/Joplin/_resources/titanic_sample_random_forest.py

#!/usr/bin/env python
# coding: utf-8

# In[4]:


import pandas as pd


# In[5]:


# 1、获取数据
titianic = pd.read_csv("./titanic.csv")
titianic.head()


# In[6]:


# 筛选特征值和目标值
x = titianic[["pclass", "age", "sex"]]
y = titianic["survived"]


# In[7]:


x.head()


# In[8]:


y.head()


# In[9]:


# 2、数据处理
# 1) 缺失值处理 age字段
x.loc[:, "age"] = x["age"].fillna(x["age"].mean())


# In[10]:


# 2) 转换成字典
x = x.to_dict(orient="records")


# In[11]:


from sklearn.model_selection import train_test_split

# 3、数据集划分
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=22)


# In[12]:


# 4、字典特征抽取
from sklearn.feature_extraction import DictVectorizer


# In[13]:


transfer = DictVectorizer()
x_train = transfer.fit_transform(x_train)
x_test = transfer.transform(x_test)


# In[14]:


from sklearn.tree import DecisionTreeClassifier
from sklearn.tree import export_graphviz

# 5、模型训练和预测
estimator = DecisionTreeClassifier(criterion="entropy")
estimator.fit(x_train, y_train)
# 6、 模型评估
# 方法一: 直接比对真实值和预测值
y_predict = estimator.predict(x_test)
print("直接比对真实值和预测值:\n", y_test == y_predict)

# 方法二: 计算准确率
score = estimator.score(x_test, y_test)
print("准确率: \n", score)

# 6) 决策树的可视化
export_graphviz(estimator, out_file="./titian_tree.dot", feature_names=transfer.get_feature_names_out())


# #### 随机森林预测泰坦尼克数据集

# In[15]:


from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import GridSearchCV


# In[16]:


# 4) 随机森林算法预估
estimator = RandomForestClassifier(n_estimators=120, random_state=22, max_features="sqrt")  # n_estimators: 树的数量
# 加入网格搜索与交叉验证
param_grid = {"n_estimators": [120, 200, 300, 500, 800, 1200],
              "max_depth": [5, 8, 15, 25, 30]}
estimator = GridSearchCV(estimator, param_grid=param_grid, cv=3)
estimator.fit(x_train, y_train)

# 5) 模型评估
# 方法一: 直接比对真实值和预测值
y_predict = estimator.predict(x_test)
print("y_predict:\n", y_predict)
print("直接比对真实值和预测值:\n", y_test == y_predict)

# 方法二: 计算准确率
score = estimator.score(x_test, y_test)
print("准确率: \n", score)

# 6. 获取最佳参数
print("最佳参数: \n", estimator.best_params_)
print("最佳结果: \n", estimator.best_score_)
print("最佳估计器: \n", estimator.best_estimator_)
print("交叉验证结果: \n", estimator.cv_results_)