Python网络爬虫-FIFA球员数据爬取
FIFA球员数据爬取
一．选题背景
世界杯的开展，给全世界广大球迷带来了狂欢和盛宴，球员们的精彩表现给世界各地的观众带来了一场场精彩绝伦的比赛。在比赛中，球员是决定比赛胜负的关键，为了探究世界球员的特点和能力，本文爬取了FIFA中的球员数据，通过对数据进行可视化探索和建模分析，能够更好的发掘影响球员能力的重要因素，从而有利于球员的成长和发现。
二．爬虫设计方案
1.爬虫名称
FIFA球员信息多线程爬虫。
2.爬取的内容和数据特征分析
本文的爬取信息为：球员名字，基本信息（出生年月，身高，体重）,Overall_Rating, Potential, Value, Wage以及ATTACKING，SKILL，MOVEMENT，POWER，MENTALITY，DEFENDING，GOALKEEPING模块下的具体信息。
数据特征分析主要在于分析球员能力的分布，并探究球员潜力（Potential）与不同能力之间的关系。
3.爬虫设计方案概述
实现思路
在球员大致信息页面中爬取球员具体信息页面网址，在通过爬取到的球员具体信息网址进入网址中爬取具体信息。
其中在主页面，也就是球员具体信息入门页面处，每个主页面含有60个球员信息入口界面，本文通过gevent库创建多线程爬虫（开启60个任务各自爬取对面副页面的球员具体信息）对球员具体信息进行爬取。除此之外，还需调用time，request，lxml和random库 随后调用numpy，pandas，seanborn，sklearn和matplotlib库对数据进行分析
技术难点
多行程并行爬虫，球员信息的定位，对于反爬进制的破解，回归方程的设立。
三．主题页面的结构特征分析
1.主题页面的结构和特征分析
球员的大致内容如下所示：
球员具体信息如下所示：
2.Htmls页面解析
每个球员的大致内容页面含有60个球员具体信息页面入口，在球员大致内容信息页面依次爬取60个具体信息 界面，并通过开启60个线程，同时爬取60个球员的具体信息。并通过xpath定位到具体的信息的地址，通过逐个查找找到需要数据的所在位置,发现所需要的数据都在
下的3.节点（标签）查找方法与遍历方法
查找方法：lxml库的xpath函数
遍历方法，人工确定数据位置
四．网络爬虫程序设计
1. 数据的爬取与采集
# 导入模块
import gevent
from gevent import monkey
monkey.patch_all()
import time
import requests
import random
from lxml import etree
import pandas as pd
class FIFA21():
def __init__(self):
self.headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64; Trident/7.0; rv:11.0) like Gecko"}
self.baseURL ='https://sofifa.com/'
self.path='data.csv'
self.row_num=0
self.title=['player_name','basic_info','Overall_rating','Potential','Value','Wage',
'preferred_foot','weak_foot','skill_moves','international_reputation']
self.small_file=['LS','ST','RS','LW','LF','CF','RF','RW','LAM','CAM','RAM','LM','LCM','CM','RCM','RM','LWB',
'LDM','CDM','RDM','RWB','LB','LCB','CB','RCB',
'RB','GK']
self.details=['Crossing','Finishing','Heading Accuracy','Short Passing','Volleys','Dribbling,Curve',
'FK Accuracy','Long Passing','Ball Control','Acceleration','Sprint Speed','Agility','Reactions',
'Balance','Shot Power','Jumping','Stamina','Strength','Long Shots','Aggression','Interceptions','Positioning',
'Vision','Penalties','Composure','Defensive' 'Awareness','Standing Tackle','Sliding Tackle','GK Diving',
'GK Handling','GK Kicking','GK Positioning','GK Reflexes']
for name in self.title:
exec ('self.'+name+'=[]')
for name in self.small_file:
exec('self.' + name + '=[]')
for i in range(len(self.details)):
exec('self.details_' +str(i) + '=[]')
def loadPage(self,url):
time.sleep(random.random())
return requests.get(url, headers=self.headers).content
def get_player_links(self,url):
content = self.loadPage(url)
html = etree.HTML(content)
player_links=html.xpath("//div[@class='card']/table[@class='table table-hover persist-area']/tbody[@class='list']"
"/tr/td[@class='col-name']/a[@role='tooltip']/@href")
result=[]
for link in player_links:
result.append(self.baseURL[:-1]+link)
return result
#return [self.baseURL[:-1]+link for link in player_links]
def next_page(self, url):
content = self.loadPage(url)
html = etree.HTML(content)
new_page= html.xpath("//div[@class='pagination']/a/@href")
if url == self.baseURL:
return self.baseURL[:-1]+new_page[0]
else:
if len(new_page)==1:
return 'stop'
else:
return self.baseURL[:-1]+new_page[1]
def Get_player_small_field(self,html):
content = html.xpath(
"//div[@class='center']/div[@class='grid']/div[@class='col col-4']/div[@class='card calculated']/div[@class='field-small']"
"/div[@class='lineup']/div[@class='grid half-spacing']/div[@class='col col-2']/div/text()")
content=content[1::2]
length=len(content)
for i in range(length):
exec ('self.'+self.small_file[i]+'.append('+'\''+content[i]+'\''+')')
#return dict(zip(keys,values))
def Get_player_basic_info(self,html):
player_name=html.xpath("//div[@class='center']/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/div[@class='info']/h1/text()")
player_basic=html.xpath("//div[@class='center']/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/div[@class='info']/div[@class='meta ellipsis']/text()")
exec ("self.player_name.append("+"\""+player_name[0]+"\""+")")
exec ('self.basic_info.append(' +'\''+player_basic[-1]+'\''+')')
def Get_rating_value_wage(self,html):
overall_potential_rating = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/span[1]/text()")
if len(overall_potential_rating)==0:
overall_potential_rating=html.xpath("//div[1]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/span[1]/text()")
value_wage = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/text()")
if len(value_wage)==0:
value_wage = html.xpath("//div[1]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/text()")
exec('self.Overall_rating.append('+'\''+overall_potential_rating[0]+'\''+')')
exec('self.Potential.append('+'\''+overall_potential_rating[1]+'\''+')')
exec('self.Value.append(' + '\'' + value_wage[2] + '\'' + ')')
exec('self.Wage.append(' + '\'' + value_wage[3] + '\'' + ')')
# / html / body / div[1] / div / div[2] / div[1] / section / div[1] / div / span
def Get_profile(self,html):
profile=html.xpath(
"//div[@class='center']/div[@class='grid']/div[@class='col col-12']/div[@class='block-quarter'][1]"
"/div[@class='card']/ul[@class='pl']/li[@class='ellipsis']/text()[1]")
exec('self.preferred_foot.append(' +'\''+ profile[0]+'\''+')')
exec('self.weak_foot.append(' +'\''+ profile[1]+'\''+')')
exec('self.skill_moves.append(' + '\'' + profile[2] + '\'' + ')')
exec('self.international_reputation.append(' + '\'' + profile[3] + '\'' + ')')
def Get_detail(self,html):
#// *[ @ id = "body"] / div[3] / div / div[2] / div[9] / div / ul / li[1] / span[1]
keys=html.xpath("//div[3]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[2]/text()")
if(len(keys)==0):
keys = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[2]/text()")
values=html.xpath("//div[3]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[1]/text()")
if (len(values)==0):
values = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[1]/text()")
values=values[:len(keys)]
values=dict(zip(keys,values))
for i in range(len(self.details)):
if self.details[i] in keys:
exec ('self.details_'+str(i)+'.append('+'\''+values[self.details[i]]+'\''+')')
else:
exec ('self.details_'+str(i)+'.append('+'\'Nan\''+')')
#return
def start_player(self,url):
content = self.loadPage(url)
html = etree.HTML(content)
#info= {}
self.Get_player_basic_info(html)
self.Get_profile(html)
self.Get_rating_value_wage(html)
self.Get_player_small_field(html)
self.Get_detail(html)
self.row_num+=1
def startWork(self):
current_url=self.baseURL
while(current_url!='stop'):
print(current_url)
player_links = self.get_player_links(current_url)
# spawn创建协程任务，并加入到任务队列里
#print(player_links)
jobs=[]
for link in player_links:
jobs.append(gevent.spawn(self.start_player, link))
#jobs = [gevent.spawn(self.start_player, link) for link in player_links]
# 父线程阻塞，等待所有任务结束后继续执行
gevent.joinall(jobs)
current_url=self.next_page(current_url)
self.save()
# 循环get队列的数据，直到队列为空则退出
def save(self):
exec ('df=pd.DataFrame()')
for name in self.title:
exec ("df["+"\'"+name+"\'"+"]=self."+name)
for name in self.small_file:
exec ('df['+'\"'+name+'\"'+']=self.'+name)
for i in range(len(self.details)):
exec ('df['+'\"'+self.details[i]+'\"'+']=self.details_'+str(i))
exec ('df.to_csv(self.path,index=False)')
if __name__ == "__main__":
start = time.time()
spider=FIFA21()
spider.startWork()
stop = time.time()
print("\n[LOG]: %f seconds..." % (stop - start))
2. 对数据进行清洗和处理
通过查阅信息知道部分爬取下来的信息不是本文分析所需要的，故将不需要的信息删除掉。
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
data=pd.read_csv('data.csv') #读入数据
data.info() #查看数据的信息
#选取自己需要的数据
left_data=data.iloc[:,:9]
right_data=data.iloc[:,37:]
data=pd.concat([left_data,right_data],axis=1)
data.drop(['basic_info','Value','Wage'],axis=1,inplace=True)
data.dropna(inplace=True) #删除缺失值
for index in data.columns[6:]:
data[index]=data[index].astype('float64')
data.info()
3. 数据分析与可视化
fig, ax= plt.subplots(nrows = 1, ncols = 2)
fig.set_size_inches(14,4)
sns.boxplot(data = data.loc[:,["Overall_rating",'Potential']], ax = ax[0])
ax[0].set_xlabel('')
ax[0].set_ylabel('')
sns.distplot(a = data.loc[:,["Overall_rating"]], bins= 10, kde = True, ax = ax[1], \
label = 'Overall_rating')
sns.distplot(a = data.loc[:,["Potential"]], bins= 10, kde = True, ax = ax[1], \
label = 'Potential')
ax[1].legend()
sns.jointplot(x='Overall_rating',y = 'Potential',data =data,kind = 'scatter')
fig.tight_layout()
查看potential和overall_rating各自的分布和关系
查看分类变量的分布
fig, ax = plt.subplots(nrows = 1, ncols = 3)
fig.set_size_inches(12,4)
sns.countplot(x = data['preferred_foot'],ax = ax[0])
sns.countplot(x = data['weak_foot'],ax = ax[1])
sns.countplot(x = data['skill_moves'],ax = ax[2])
fig.tight_layout()
两个变量之间散点图分析
通过分析相关系数图，本文发现和Potential变量相关的变量有Overall_rating, Reactions和shot power等，做出散点图如下。
sns.jointplot(x='Reactions',y = 'Potential',data =data,kind = 'scatter')
sns.jointplot(x='Shot Power',y = 'Potential',data =data,kind = 'scatter')
sns.jointplot(x='Vision',y = 'Potential',data =data,kind = 'scatter')
4. 建立回归模型
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
x=df.drop('Potential',axis=1)
y=df['Potential']
x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.3)
model=LinearRegression()
model.fit(x_train,y_train)
from sklearn.metrics import r2_score
r2_score(y_test,model.predict(x_test))
模型最终的R方为0.50，效果较好。
5. 数据持久化
df.to_csv('df.csv')
保存为csv文件
6. 源码
爬虫
# 导入模块
import gevent
from gevent import monkey
monkey.patch_all()
import time
import requests
import random
from lxml import etree
import pandas as pd
class FIFA21():
def __init__(self):
self.headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64; Trident/7.0; rv:11.0) like Gecko"}
self.baseURL ='https://sofifa.com/'
self.path='data.csv'
self.row_num=0
self.title=['player_name','basic_info','Overall_rating','Potential','Value','Wage',
'preferred_foot','weak_foot','skill_moves','international_reputation']
self.small_file=['LS','ST','RS','LW','LF','CF','RF','RW','LAM','CAM','RAM','LM','LCM','CM','RCM','RM','LWB',
'LDM','CDM','RDM','RWB','LB','LCB','CB','RCB',
'RB','GK']
self.details=['Crossing','Finishing','Heading Accuracy','Short Passing','Volleys','Dribbling,Curve',
'FK Accuracy','Long Passing','Ball Control','Acceleration','Sprint Speed','Agility','Reactions',
'Balance','Shot Power','Jumping','Stamina','Strength','Long Shots','Aggression','Interceptions','Positioning',
'Vision','Penalties','Composure','Defensive' 'Awareness','Standing Tackle','Sliding Tackle','GK Diving',
'GK Handling','GK Kicking','GK Positioning','GK Reflexes']
for name in self.title:
exec ('self.'+name+'=[]')
for name in self.small_file:
exec('self.' + name + '=[]')
for i in range(len(self.details)):
exec('self.details_' +str(i) + '=[]')
def loadPage(self,url):
time.sleep(random.random())
return requests.get(url, headers=self.headers).content
def get_player_links(self,url):
content = self.loadPage(url)
html = etree.HTML(content)
player_links=html.xpath("//div[@class='card']/table[@class='table table-hover persist-area']/tbody[@class='list']"
"/tr/td[@class='col-name']/a[@role='tooltip']/@href")
result=[]
for link in player_links:
result.append(self.baseURL[:-1]+link)
return result
#return [self.baseURL[:-1]+link for link in player_links]
def next_page(self, url):
content = self.loadPage(url)
html = etree.HTML(content)
new_page= html.xpath("//div[@class='pagination']/a/@href")
if url == self.baseURL:
return self.baseURL[:-1]+new_page[0]
else:
if len(new_page)==1:
return 'stop'
else:
return self.baseURL[:-1]+new_page[1]
def Get_player_small_field(self,html):
content = html.xpath(
"//div[@class='center']/div[@class='grid']/div[@class='col col-4']/div[@class='card calculated']/div[@class='field-small']"
"/div[@class='lineup']/div[@class='grid half-spacing']/div[@class='col col-2']/div/text()")
content=content[1::2]
length=len(content)
for i in range(length):
exec ('self.'+self.small_file[i]+'.append('+'\''+content[i]+'\''+')')
#return dict(zip(keys,values))
def Get_player_basic_info(self,html):
player_name=html.xpath("//div[@class='center']/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/div[@class='info']/h1/text()")
player_basic=html.xpath("//div[@class='center']/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/div[@class='info']/div[@class='meta ellipsis']/text()")
exec ("self.player_name.append("+"\""+player_name[0]+"\""+")")
exec ('self.basic_info.append(' +'\''+player_basic[-1]+'\''+')')
def Get_rating_value_wage(self,html):
overall_potential_rating = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/span[1]/text()")
if len(overall_potential_rating)==0:
overall_potential_rating=html.xpath("//div[1]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/span[1]/text()")
value_wage = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/text()")
if len(value_wage)==0:
value_wage = html.xpath("//div[1]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='bp3-card player']/section[@class='card spacing']/div[@class='block-quarter']/div/text()")
exec('self.Overall_rating.append('+'\''+overall_potential_rating[0]+'\''+')')
exec('self.Potential.append('+'\''+overall_potential_rating[1]+'\''+')')
exec('self.Value.append(' + '\'' + value_wage[2] + '\'' + ')')
exec('self.Wage.append(' + '\'' + value_wage[3] + '\'' + ')')
# / html / body / div[1] / div / div[2] / div[1] / section / div[1] / div / span
def Get_profile(self,html):
profile=html.xpath(
"//div[@class='center']/div[@class='grid']/div[@class='col col-12']/div[@class='block-quarter'][1]"
"/div[@class='card']/ul[@class='pl']/li[@class='ellipsis']/text()[1]")
exec('self.preferred_foot.append(' +'\''+ profile[0]+'\''+')')
exec('self.weak_foot.append(' +'\''+ profile[1]+'\''+')')
exec('self.skill_moves.append(' + '\'' + profile[2] + '\'' + ')')
exec('self.international_reputation.append(' + '\'' + profile[3] + '\'' + ')')
def Get_detail(self,html):
#// *[ @ id = "body"] / div[3] / div / div[2] / div[9] / div / ul / li[1] / span[1]
keys=html.xpath("//div[3]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[2]/text()")
if(len(keys)==0):
keys = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[2]/text()")
values=html.xpath("//div[3]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[1]/text()")
if (len(values)==0):
values = html.xpath("//div[2]/div[@class='grid']/div[@class='col col-12']"
"/div[@class='block-quarter']/div[@class='card']/ul[@class='pl']/li/span[1]/text()")
values=values[:len(keys)]
values=dict(zip(keys,values))
for i in range(len(self.details)):
if self.details[i] in keys:
exec ('self.details_'+str(i)+'.append('+'\''+values[self.details[i]]+'\''+')')
else:
exec ('self.details_'+str(i)+'.append('+'\'Nan\''+')')
#return
def start_player(self,url):
content = self.loadPage(url)
html = etree.HTML(content)
#info= {}
self.Get_player_basic_info(html)
self.Get_profile(html)
self.Get_rating_value_wage(html)
self.Get_player_small_field(html)
self.Get_detail(html)
self.row_num+=1
def startWork(self):
current_url=self.baseURL
while(current_url!='stop'):
print(current_url)
player_links = self.get_player_links(current_url)
# spawn创建协程任务，并加入到任务队列里
#print(player_links)
jobs=[]
for link in player_links:
jobs.append(gevent.spawn(self.start_player, link))
#jobs = [gevent.spawn(self.start_player, link) for link in player_links]
# 父线程阻塞，等待所有任务结束后继续执行
gevent.joinall(jobs)
current_url=self.next_page(current_url)
self.save()
# 循环get队列的数据，直到队列为空则退出
def save(self):
exec ('df=pd.DataFrame()')
for name in self.title:
exec ("df["+"\'"+name+"\'"+"]=self."+name)
for name in self.small_file:
exec ('df['+'\"'+name+'\"'+']=self.'+name)
for i in range(len(self.details)):
exec ('df['+'\"'+self.details[i]+'\"'+']=self.details_'+str(i))
exec ('df.to_csv(self.path,index=False)')
if __name__ == "__main__":
start = time.time()
spider=FIFA21()
spider.startWork()
stop = time.time()
print("\n[LOG]: %f seconds..." % (stop - start))
数据分析和建模
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
data=pd.read_csv('data.csv')
left_data=data.iloc[:,:9]
right_data=data.iloc[:,37:]
data=pd.concat([left_data,right_data],axis=1)
data.drop(['basic_info','Value','Wage'],axis=1,inplace=True)
data.dropna(inplace=True)
for index in data.columns[6:]:
data[index]=data[index].astype('float64')
data.info()
fig, ax= plt.subplots(nrows = 1, ncols = 2)
fig.set_size_inches(14,4)
sns.boxplot(data = data.loc[:,["Overall_rating",'Potential']], ax = ax[0])
ax[0].set_xlabel('')
ax[0].set_ylabel('')
sns.distplot(a = data.loc[:,["Overall_rating"]], bins= 10, kde = True, ax = ax[1], \
label = 'Overall_rating')
sns.distplot(a = data.loc[:,["Potential"]], bins= 10, kde = True, ax = ax[1], \
label = 'Potential')
ax[1].legend()
sns.jointplot(x='Overall_rating',y = 'Potential',data =data,kind = 'scatter')
fig.tight_layout()
fig, ax = plt.subplots(nrows = 1, ncols = 3)
fig.set_size_inches(12,4)
sns.countplot(x = data['preferred_foot'],ax = ax[0])
sns.countplot(x = data['weak_foot'],ax = ax[1])
sns.countplot(x = data['skill_moves'],ax = ax[2])
fig.tight_layout()
corr2 = data.select_dtypes(include =['float64','int64']).\
loc[:,data.select_dtypes(include =['float64','int64']).columns[:]].corr()
fig,ax = plt.subplots(nrows = 1,ncols = 1)
fig.set_size_inches(w=24,h=24)
sns.heatmap(corr2,annot = True,linewidths=0.5,ax = ax)
plt.savefig('1.png')
sns.jointplot(x='Reactions',y = 'Potential',data =data,kind = 'scatter')
sns.jointplot(x='Shot Power',y = 'Potential',data =data,kind = 'scatter')
sns.jointplot(x='Vision',y = 'Potential',data =data,kind = 'scatter')
sns.jointplot(x='Composure',y = 'Potential',data =data,kind = 'scatter')
df=data[['Potential','Overall_rating','Short Passing','Long Passing','Ball Control','Reactions','Shot Power','Vision','Composure']]
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
x=df.drop('Potential',axis=1)
y=df['Potential']
x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.3)
model=LinearRegression()
model.fit(x_train,y_train)
from sklearn.metrics import r2_score
r2_score(y_test,model.predict(x_test))
df.to_csv('df.csv')
五．总结
在完成爬虫课程设计的过程中，会遇到很多问题，如在爬取过程中的防爬机制，打代码时不断的出错，本文需要主动的上网去找解决问题的办法，同时由于数据量大，时间长，本文更改了爬取的方式，使用多线程来对数据进行爬取。爬取数据时由于网页数据安排并不是特别简洁，故本文需要对网页数据的爬取进行一个一个的手动定位，比较麻烦。在数据分析时爬取的数据并非你全部都是本文后续分析的需要，故需要对数据进行筛选，同时数据存在一定量的缺失值，故本文对数据删除了缺失值。
达到了预期的目标与收获：
本次课程设计，让作者学会了爬虫的基本原理，数据清洗和处理的方式。其中，爬虫常用requests模拟请求网页，获取数据。同时深化了作者对于Sklearn库使用的能力，直接调用SKlearn库来对数据进行建模分析会使得效率更高。
需要改进的建议：
在进行线性回归时，加入更多数据使数据更具有预测性。
NBA常规赛 火箭106-111小牛视频集锦
芦苇谈伦纳德