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

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

# In[78]:


import pandas as pd

# In[79]:


# 1 获取数据
data = pd.read_csv("./FBlocation/train.csv")

# In[80]:


data.head()

# In[81]:


# 2 基本数据处理
# 1) 缩小数据处理
data = data.query("x > 1.5 & x < 2.5 & y > 1.0 & y < 1.5")

# In[82]:


data.head()

# In[83]:


# 2) 处理时间特征
time_value = pd.to_datetime(data["time"], unit="s")

# In[84]:


date = pd.DatetimeIndex(time_value)

# In[85]:

data["day"] = date.day
data["weekday"] = date.weekday
data["hour"] = date.hour

# In[86]:


data.head()

# In[87]:


# 3) 过滤签到次数少的地点
place_count = data.groupby("place_id").count()["row_id"]

# In[88]:


place_count[place_count > 3].head()

# In[89]:


data_final = data[data["place_id"].isin(place_count[place_count > 3].index.values)]

# In[90]:


# 4) 筛选特征值和目标值
x = data_final[["x", "y", "accuracy", "day", "weekday", "hour"]]
y = data_final["place_id"]

# In[91]:


x.head()

# In[92]:


y.head()

# In[93]:


# 5) 数据集划分
from sklearn.model_selection import train_test_split

# In[94]:


x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=22)
# 数据处理完毕


# In[95]:


from sklearn.preprocessing import StandardScaler
from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import GridSearchCV

# In[96]:


# 3. 特征工程 标准化
transfer = StandardScaler()
x_train = transfer.fit_transform(x_train)  # fit_transform训练集
x_test = transfer.transform(x_test)  # transform测试集

# 4. KNN 算法预估
estimator = KNeighborsClassifier()

# 加入网格搜索与交叉验证
# 参数准备
param_grid = {"n_neighbors": [3, 5, 7, 9]}
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_)