包钢无缝管,大哥大电影,3u8506
本文实例为大家分享了python实现txt文件格式转换为arff格式的具体代码，供大家参考，具体内容如下
将文件读取出来的时候默认都是字符型的，所以有转换出来有点问题，但是还是可以用的。
文件要求第一行是你对应的属性名，之后是数字。
我要评论
import sys
import re
relationname = ""
filename = ""
if (len(sys.argv)<2):
print("usage:\npython arff.py myrelationname filename.txt")
else:
relationname = sys.argv[1]
filename = sys.argv[2]
class arff:
def __init__(self, r, f):
self.relationname = r if r is not "" else "machinelearning"
f = f if f is not "" else "mmg_data.txt"
self.file1 = open(f, 'r')
self.data = []
self.names = []
self.parsedata()
self.writetofile()
def parsedata(self):
firstline = true
for line in self.file1.readlines():
if not firstline:
try:
line = line.replace("\n", "")
words = line.split(" ")
except valueerror:
print("cant parse file!!")
self.data.append(words)
else:
firstline = false
line = line.replace("\n", "")
words = line.split(" ")
self.names = words
def gettype(self, value):
v = ""
if(type(value) == type(1)):
v = "numeric"
elif(type(value) == type(1.0)):
v = "numeric"
elif(re.match("[0-9]{4}\-[0-9]{2}\-[0-9]{2}\s[0-9]{2}\:[0-9]{2}\:[0-9]{2}", value)):
v = "date " + "yyyy-mm-dd hh:mm:ss"
elif(type(value) == type("string")):
v = "string"
elif(v == ""):
print("data type "+value+" not supported yet.")
return v
def writetofile(self):
values = self.data[0]
file2 = open("dexhunter_test_result.arff", 'w+' )
self.relationname+="\n"
relationstring = '@relation ' + self.relationname
file2.write(''+relationstring+'')
for i in range(len(self.names)):
str2 = "@attribute " + self.names[i] + " " + self.gettype( values[i] ) + "\n"
file2.write(''+str2+'')
file2.write('''''@data\n''')
for line in self.data:
try:
file2.write(",".join(line)+"\n")
except unicodeencodeerror:
print("cant write data to file!!")
arff(relationname, filename)
以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持移动技术网。
如对本文有疑问，请在下面进行留言讨论，广大热心网友会与你互动！！ 点击进行留言回复
Python 实现将numpy中的nan和inf,nan替换成对应的均值
python爬虫把url链接编码成gbk2312格式过程解析
网友评论