又到了中文+Python基础的时刻了，多开心多轻松的时刻，It's show time! 伴随着这首刺破耳膜的歌，指尖起舞。。。（---你是在弹Piano吗？ ---我在编程 ---She精病）
推导式
[i**2 for i in range(10)]#也可以用for啥啥，啥啥啥循环，但这个更优雅和快速
[0, 1, 4, 9, 16, 25, 36, 49, 64, 81]
装饰器-不改变原函数的情况下更新输出
def mse_box(list_input): """ input: list output: int, mean square error of the list """ mean_list = sum(list_input)/len(list_input) list_square_error = [(i-mean_list)**2 for i in list_input] mean_square_error = sum(list_square_error)/len(list_square_error) return(mean_square_error)mse_box([1, 2, 3, 2])
0.5
放出装饰器看看
def outer(func):#参数为函数func，接下来func会放mse_box() def decorate(*args, **kwargs):#*args, **kwargs代表可以接收各类参数了 print("Input list is: {0}".format(*args)) input0 = list(*args)#不能直接print(*args)，要加list() print("Function name is: {}, output is:".format(func.__name__)) print (func(*args, **kwargs)) mse0 = func(*args, **kwargs) print("make adjustment to {0}:".format(func.__name__)) print("each element of input decreases mse, and the new output is: ") print([i-mse0 for i in input0])#如果要输出该部分，加个return()即可 return decorate@outerdef mse_box(list_input): """ input: list output: int, mean square error of the list """ mean_list = sum(list_input)/len(list_input) list_square_error = [(i-mean_list)**2 for i in list_input] mean_square_error = sum(list_square_error)/len(list_square_error) return(mean_square_error)mse_box([1, 2, 3, 2]) 
装饰器的作用：此例中，为了修改mse_box函数，单独在外面将mse_box该func及其参数拎出来，在外头修改。
单独修改的好处：假如第二个模块引用了mse_box，且该模块只需要计算均方差，不需要进行任何改动，那装饰器就make a difference了。
Dict
keys = ["a", "b", "c"]value = [1, 2, 3]dict(zip(keys, value))#用dict嵌套zip可以比较快地生成字典
{'a': 1, 'b': 2, 'c': 3}
同样地，用推导式将数据转换成字典比较方便，但有些费脑
raw_data = "水果：柿子，蔬菜：西红柿，其他：灯笼"{a:b for a,b in (j.split("：") for j in [i for i in raw_data.split("，")])} {'其他': '灯笼', '水果': '柿子', '蔬菜': '西红柿'}
try_except
使用try可以避免报错就中断整个程序，使用except来发现是不是我们注释的地方中断了，方便找错
try: print(raw_data[100])except: print("If you see this, raw_data doesn't have length as long as 100")try: print(raw_data[:2] == "水果")except: print("If you see this, first two words of raw_data are not 水果")try: print(raw_data+1)except: print("If you see this, raw_data can't add number") If you see this, raw_data doesn't have length as long as 100
True If you see this, raw_data can't add numberOS文档操作
import os#自带包，不需要额外pip install xxx#不清楚os有什么功能，输入os.摁TAB吧os.getcwd()# get current working directoryos.listdir()[-5:]#list directory of current working directory,输出格式为list#查看所有Jupyter notebook格式文档[file for file in os.listdir() if file[-5:] == 'ipynb']#推导式，怎么又是你# 改变目前工作路径os.chdir("/code/python_data_analysis") 正则表达式及re包的运用
Ans:
- [c,m,f]an: can, man, fan; [^b]og to skip bog
- [^a-z]\w+ skip lower case begined string; \w means [A-Za-z0-9]; \d means [0-9]
- z{3} match z three times: uzzz, wuzzzz; .{2,6} match string with length of 2-6
- ? match ?
- whitespace: space (␣), the tab (\t), the new line (\n) and the carriage return (\r)
- \s will match any of the specific whitespaces above
- \D represents any non-digit character, \S any non-whitespace character, and \W any non-alphanumeric
- ^Mission: successful$ ^为字符串开始 and $为字符串结尾
- ^(file_\w+) can match file_record_transcript in file_record_transcript.pdf
- ^([A-Z]\w{2} (\d{4})) 括号中为提取的信息，此处不但提取Jan 1987，还提取1987
- ^I love cats|I love dogs$ match "I love cats"或"I love dogs"
- ^The.* match string starting with "The"
正则表达式的练习在线网址: https://regexone.com/
触发支线任务--爬虫
dir_name = os.listdir()dir_name[:5]import refor i in [re.findall("csv_\d{1,2}\.csv", i) for i in dir_name]: if i != []: print(i[0]) 
os.rename("Untitled Folder 1", "New Folder 1")#可以重命名 文档输入与输出
import pandas as pd#经常用到的一个包，装了anaconda就自带df1 = pd.read_csv("csv1.csv")#这个文档自己随意创建的#也可以试试pd.read_excel("表格1.xlsx"); 输入pd.read_摁TAB可以查看其他输入df1.head() 
df1_head = df1.head()df1_head.to_csv("csv1_top5.csv")#输出到文档，也可以试试df_data.to_excel("target.xlsxa)#设置index=False可以不打印额外的index#接着，制造一些数据留给你做Homework吧for i in range(10): df = df1.head(i+1) df.to_csv("csv_{}.csv".format(i+1))#也可以试试输出为excel格式(.xlsx) Homework
背景：有没有遇到这样傻逼又无奈的事情：打开一堆Excel，然后把指定位置的数据摘出来放个新的表格。。。考验你百度(Google)的能力了
例如，将csv_n.csv打开（n为1到10），然后分别要"csv_n.csv"的第n行数据