python编程规则与技巧-（持续记录）

一、头文件
基本模块导入

import pandas as pd
import numpy as np
import cv2
import os
import matplotlib.pyplot as plt
import re

机器学习模块：

from sklearn.cross_validation import train_test_split  #分数据
import mglearn 
from sklearn.linear_model import Ridge 
from sklearn import datasets, linear_model
from sklearn.linear_model import LogisticRegression
from sklearn.svm import LinearSVC
from mglearn import plot_2d_separator
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_absolute_error

深度学习模块：

from keras.models import Sequential
from keras.layers import Dense,Dropout,Flatten,Conv2D,MaxPooling2D
from keras.models import load_model

读取文件方法，好多包都可读取文件：
C:/pypractise/fangzhen2/cj.csv

text = open('test_txt/14.txt','r')
q = text.read()

data=pd.read_csv('C:/pypractise/3/Advertising.csv')

读取文件列表，以便在依次遍历文件

def get_imglist(path):
    return [os.path.join(path,f) for f in os.listdir(path) if f.endswith('.jpg')]

写入exls 大文件可用

def data_w(datas):
    f= xlsxwriter.Workbook('demo1.xlsx')#创建一个excel文件
    sheet1= f.add_worksheet(u'sheet1')
    i = 0
    j = 0
    for data in datas[0]:        
        sheet1.write(i,j,data)
        i=i+2
    i = 1
    j = 0
    for data1 in datas[1]:
        sheet1.write(i,j,data1)
        i=i+2
    f.close()

将文件写进行excel表格中：

def data_write(datas):
    f = xlwt.Workbook()
    sheet1 = f.add_sheet(u'ex.xls',cell_overwrite_ok=True) #创建sheet

    #将数据写入第 i 行，第 j 列
    i = 0    #行
    j = 0   #列
    for data in datas:        
        sheet1.write(i,j,data)
        i=i+1
#    i=1
#    for data1 in datas[1]:
#        sheet1.write(i,j,data1)
#        i=i+2
    f.save('ex.xls') #保存文件

将文件txt的写到csv中；

def data_write():
    '''
    函数为将txt数据写入到csv中
    '''    
    with open('shuju.csv', 'w+',newline='') as csvfile:
        spamwriter = csv.writer(csvfile, dialect='excel')
        # 读要转换的txt文件，文件每行各词间以@@@字符分隔
        with open('che.txt', 'r',encoding='utf-8') as filein:
            for line in filein:
                line_list = line.strip('\n').split(' ')
                spamwriter.writerow(line_list)

正则语句切分. 文献去头尾;

#(\.(?=\s+(?:[A-Z])))|(\). )|(\.[ 0-9|A-Z])|(ACKNOWLEDGMENTS.*$)|(.*AUTHOR INFORMATION*$)|(REFERENCES.*$)|(References.*$)|(\?+)|(^.*ABSTRACT:)|(^.*INTRODUCTION )| (^.*CONSPECTUS:)
#正则规则去除开头结尾和*？
# 切分 \.(?=[\d+](?=\s+(?:[A-Z])))|\.(?=\s+(?:[A-Z])))
# 删除开头结尾 (ACKNOWLEDGMENTS.*$)|(AUTHOR.INFORMATION.*$)|(REFERENCES.*$)|(References.*$)|(^.*ABSTRACT:)|(^.*INTRODUCTION )| (^.*CONSPECTUS:)|(ASSOCIATED.CONTENT.*$)
#替换 ？和* (\?+)|(\*+)
re.sub(pattern1,'',q)
re.compile(r' ')

调用API翻译 百度：
ID自己的号

def fanyi(yuju):
   
    '''
    出入Q 返回src 和dst
    '''
    
    myurl = '/api/trans/vip/fieldtranslate'   #分网页
    appid = '    '
    secretKey =  '   '
    
    q = yuju
    q.encode(encoding = 'utf-8')
    salt = random.randint(32768, 65536)
    salt = str(salt)
    domain = 'medicine'
    sign = appid + q + salt + domain + secretKey   #签名
    
    m1 = hashlib.md5()  #创建一个md5加密模式的hash对象
    m1.update(sign.encode(encoding = 'utf-8'))  #跟新对象
    sign = m1.hexdigest()  #返回摘要
    
    fromLang = 'en'
    toLang = 'zh'
    httpClient = None
    
    myurl = myurl+'?q='+parse.quote(q)+'&from='+fromLang+'&to='+toLang+'&appid='+appid+'&salt='+salt+'&domain='+'medicine'+'&sign='+sign 

    try:
        httpClient = http.client.HTTPConnection('api.fanyi.baidu.com')    #主网页
        httpClient.request('GET', myurl)
        response = httpClient.getresponse()
        html = response.read().decode('utf-8')
        html = json.loads(html)    #返回翻译结果
#        ds1t = html["trans_result"][0]["dst"]
#        src1 = html["trans_result"][0]["src"]
    except Exception as e:
        print(e)
    finally:
        if httpClient:
            httpClient.close()
    return html

画图方法：
一般的画出来：

plt.title("ATP curve")     #设置标题
plt.xlim(xmax=2500,xmin=-250)    #设置x y轴的大小
plt.ylim(ymax=220,ymin=0)
plt.figure()
plt.plot(range(len(y_pred)),y_pred,'b',label='predict')
plt.plot(range(len(y_pred)),y_test,'r',label='test')
plt.legend(loc="upper right")    #图标的位置
plt.xlabel("the number of sales")
plt.ylabel("value of sales")
plt.show()