探索白葡萄酒的品质

# -*- coding: utf8 -*-
"""
原因: Python2.x的解释器只认识ASCII编码的py文件
解决: =>首行注释告诉解释器文件编码, #coding=utf-8
        =>Python3.x
"""
import csv
#1.读取阿比葡萄酒品质的数据
with open('white_wine.csv',mode='r') as f:
    reader = csv.reader(f)
    content=[]
    for row in reader:
        content.append(row)
#print(content[:5])
#2.查看白葡萄酒中共分为几种品质
"""
品质quality变量在数据中是一个离散变量，而不是连续的，所以它只会有固定的几个等级。那么我们用Python中自带的集合set来查看白葡萄酒中总共的品质等级
"""
qualities=[]
for item in content[1:]:
    qualities.append(int(item[-1]))
#qualities 含有多个相同的数，因此我们可以用集合set的无序性

unity_quality = set(qualities)#unity_quality是set([3, 4, 5, 6, 7, 8, 9])
print unity_quality

#3.将数据切分
"""
将数据按白葡萄酒等级`quality`进行切分为七个子集，保存到一个字典中，字典的键为`quality`具体数值，值为归属于该`quality`的样本列表
"""
content_dict={}
for row in content[1:]:
    quality = int(row[-1])
    if quality not in content_dict.keys():
        content_dict[quality]=[row]
    else:
        content_dict[quality].append(row)
print content_dict[3]

#统计在每个品质的样本量
number_tuple=[]
for quality in content_dict.keys():
    number_tuple.append((quality,len(content_dict[quality])))
#求每个数据集中fixed acidity的均值
mean_tuple=[]
for quality,samples in content_dict.items():
    sum_=0
    for sample in samples:
        sum_+=float(sample[0])
    mean_tuple.append((quality,sum_/len(content_dict[quality])))
print mean_tuple

对字典的操作不够秀。。。。。需要学习