大作业

一、boston房价预测

---

from sklearn.datasets import load_boston
boston = load_boston()
boston.keys()

#划分训练集与测试集
from sklearn.model_selection import train_test_split
x_train,x_test,y_train,y_test=train_test_split(boston.data,boston.target,test_size=0.3) 
print(x_train.shape,y_train.shape)

#一元线性回归模型
from sklearn.linear_model import LinearRegression
LineR = LinearRegression()
LineR.fit(x_train,y_train)
print("系数:",LineR.coef_)       #系数
print("截距:",LineR.intercept_)        #截距

#检测模型好坏
from sklearn.metrics import regression
y_pred = LineR.predict(x_test)   #预测
n = regression.mean_squared_error(y_test,y_pred) #预测模型均方误差
print("线性回归均方误差：",n)
a = regression.mean_absolute_error(y_test,y_pred)
print("线性回归平均绝对误差",a)
s = LineR.score(x_test,y_test)  #模型分数
print("线性回归模型分数：",s)

#多元多项式回归模型
from sklearn.preprocessing import PolynomialFeatures

poly = PolynomialFeatures(degree = 2)
x_poly_train= poly.fit_transform(x_train) #多项式化
x_poly_test  = poly.transform(x_test)

#建立模型
LineR2 = LinearRegression()
LineR2.fit(x_poly_train,y_train)


y_pred2 =LineR2.predict(x_poly_test)#预测

#检测模型好坏
n2 = regression.mean_squared_error(y_test,y_pred2)
print("多项式回归均方误差：",n2)
a2 = regression.mean_absolute_error(y_test,y_pred2)
print("多项式回归平均绝对误差",a2)
s2 = LineR2.score(x_poly_test,y_test)
print("多项式回归模型分数：",s2)

#比较线性模型与非线性模型的性能，并说明原因
非线性模型功能强，因为多项式模型是曲线，更符合与样本点的分布，而且多项式回归模型误差更小。

 二、中文文本分类

import os
import numpy as np
import sys
from datetime import datetime
import gc
path = 'E:\258'
list = os.listdir(path)
list

def list_all_files(rootdir):
    import os
    _files = []
    list = os.listdir(rootdir) #列出文件夹下所有的目录与文件
    for i in range(0,len(list)):
           path = os.path.join(rootdir,list[i])
           if os.path.isdir(path):
              _files.extend(list_all_files(path))
           if os.path.isfile(path):
              _files.append(path)
    return _files

_fs = list_all_files('E:\258')
#所有文件遍历出来

_fs

!pip install jieba
import jieba
with open(r'E:\stopsCN.txt',encoding='utf-8') as f:
    stopwords = f.read().split('
')

import jieba
def processing(tokens):
    
    tokens = "".join([char for char in tokens if char.isalpha()])# 去掉非字母汉字的字符
    
    tokens = [token for token in jieba.cut(tokens,cut_all=True) if len(token) >=2]# 结巴分词
    
    tokens = " ".join([token for token in tokens if token not in stopwords])# 去掉停用词
    return tokens

tokenList = []
targetList = []

for root,dirs,files in os.walk(path):#拼接路径，再打开文件
    for f in files:
        filePath = os.path.join(root,f)
        with open(filePath,encoding='utf-8') as f:
            content = f.read()
            
            
            target = filePath.split('\')[-2]
            targetList.append(target)
            tokenList.append(processing(content))

# 划分训练集测试集
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import GaussianNB,MultinomialNB
from sklearn.model_selection import cross_val_score
from sklearn.metrics import classification_report

x_train,x_test,y_train,y_test = train_test_split(tokenList,targetList,test_size=0.3,stratify=targetList)

vectorizer = TfidfVectorizer()
X_train = vectorizer.fit_transform(x_train)
X_test = vectorizer.transform(x_test)

#多项式朴素贝叶斯
from sklearn.naive_bayes import MultinomialNB
mnb = MultinomialNB()
module = mnb.fit(X_train,y_train)

y_predict = module.predict(X_test)#预测
scores=cross_val_score(mnb,X_test,y_test,cv=5)
print("Accuracy:%.3f"%scores.mean())
print("classification_report:
",classification_report(y_predict,y_test))

targetList.append(target)
print(targetList[0:10])
tokenList.append(processing(content))
tokenList[0:10]

# 统计测试集和预测集的各类新闻个数
testCount = collections.Counter(y_test)
predCount = collections.Counter(y_predict)
print('实际：',testCount,'
', '预测', predCount)

# 建立标签列表，实际结果列表，预测结果列表
nameList = list(testCount.keys())
testList = list(testCount.values())
predictList = list(predCount.values())
x = list(range(len(nameList)))
print("新闻类别：",nameList,'
',"实际：",testList,'
',"预测：",predictList)