1、三国演义Top10人物分析
import jieba
from wordcloud import WordCloud
import imageio
# 1.读取小说内容
with open('./novel/threekingdom.txt', 'r', encoding='utf-8') as f:
words = f.read()
counts = {} # counts = {'姓名':出现频率}
excludes = {"将军", "却说", "丞相", "二人", "不可", "荆州", "不能", "如此", "商议",
"如何", "主公", "军士", "军马", "左右", "次日", "引兵", "大喜", "天下",
"东吴", "于是", "今日", "不敢", "魏兵", "陛下", "都督", "人马", "不知",
"孔明曰", "玄德曰", "刘备", "云长"}
# 2.分词
words_list = jieba.lcut(words)
print(words_list)
for word in words_list:
if len(word) <= 1:
continue
else:
# 更新字典中的值
# counts[word] = 取出字典中原来键对应的值 ＋ 1
# counts[word] = counts[word] + 1
# 字典.get(k) 如果字典中没有这个键 ，(返回NONE)添加一个默认值：0
counts[word] = counts.get(word, 0) + 1
print(counts)
# 3.词语过滤，删除无关词，重复词
counts['孔明'] = counts['孔明'] + counts['孔明曰']
counts['玄德'] = counts['玄德'] + counts['玄德曰'] + counts['刘备']
counts['关公'] = counts['关公'] + counts['云长']
for word in excludes:
del counts[word]
# 4.排序[(), ()]
items = list(counts.items())
print(items)
# def sort_by_count(x):
# return x[1]
# items.sort(key=sort_by_count, reverse=True)
# 用列表解析排序
items.sort(key=lambda x: x[1], reverse=True)
# print(items)
li = [] # ['孔明', '孔明', '孔明',..., '曹操'...]
for i in range(10):
# 序列解包
role, count = items[i]
print(role, count)
# _ 是告诉看代码的人，循环里面不需要使用临时变量
for _ in range(count):
li.append(role)
# 得出结论
# 绘制中文词云，在WordCloud()里面设置参数
text = ' '.join(li)
wc = WordCloud(
font_path='msyh.ttc',
background_color='white',
width=800,
height=600,
# 相邻两个重复词之间的匹配，关掉
collocations=False
).generate(text)
wc.to_file('三国TOP10人物词云.png')
Top10人物
2、匿名函数 lambda
- 结构： lambda x1, x2, ...xn: 表达式
eg：
sum_num = lambda x1, x2: x1+x2
print(sum_num(2, 3))
# 结果：5
- 参数可以是无限多个，但是表达式只有一个
eg1：从大到小排序
name_info_list = [
('张三', 4500),
('李四', 9500),
('王五', 2000),
('赵六', 5500),
]
name_info_list.sort(key=lambda x: x[1], reverse=True)
print(name_info_list)
# 结果：[('李四', 9500), ('赵六', 5500), ('张三', 4500), ('王五', 2000)]
eg2：从小到大排序
stu_info = [
{"name": 'zhangsan', "age": 18},
{"name": 'lisi', "age": 30},
{"name": 'wangwu', "age": 99},
{"name": 'tianqi', "age": 3},
]
stu_info.sort(key=lambda i: i['age'])
print(stu_info)
# 结果：[{'name': 'tianqi', 'age': 3}, {'name': 'zhangsan', 'age': 18}, {'name': 'lisi', 'age': 30}, {'name': 'wangwu', 'age': 99}]
3、列表推导式，列表解析和字典解析
1、列表推导式
- 之前用普通for 创建列表
li = []
for i in range(10):
li.append(i)
print(li)
# 结果：[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
- 使用列表推导式创建列表
结构：[表达式 for 临时变量 in 可迭代对象 可以追加条件]
print([i for i in range(10)])
# 结果：[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
2、 列表解析
- 普通筛选出列表中所有的偶数
# 筛选出列表中所有的偶数
li = []
for i in range(10):
if i % 2 == 0:
li.append(i)
print(li)
# 结果：[0, 2, 4, 6, 8]
- 使用列表解析筛选偶数
print([i for i in range(10) if i % 2 == 0])
# 结果：[0, 2, 4, 6, 8]
- 筛选出列表中 大于0 的数
# 随机生成(-10, 10)的10个数
from random import randint
num_list = [randint(-10, 10) for _ in range(10)]
print(num_list)
# 输出num_list中 大于0 的数
print([i for i in num_list if i > 0])
# 结果：[-5, -1, -4, -8, -8, -1, -8, 7, 10, -4]
# [7, 10]
3、字典解析
from random import randint
stu_grades = {'student{}'.format(i): randint(50, 100) for i in range(1, 101)}
print(stu_grades)
# 筛选 大于60 分的所有学生
print({k: v for k, v in stu_grades.items() if v > 60})
结果
4、Matplotlib
- 导入
from matplotlib import pyplot as plt
- 使用100个点 绘制[0, 2π]正余弦曲线图
from matplotlib import pyplot as plt
import numpy as np
# 处理中文乱码
plt.rcParams["font.sans-serif"] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
# 使用100个点 绘制[0, 2π]正弦曲线图
# .linspace 左闭右闭区间的等差数列
x = np.linspace(0, 2*np.pi, num=100)
print(x)
y = np.sin(x)
# 正弦和余弦在同一坐标系下
cosy = np.cos(x)
plt.plot(x, y, color='g', linestyle='--')
plt.plot(x, cosy, color='r')
plt.xlabel('时间（s）')
plt.ylabel('电压（v）')
plt.title('欢迎来到python世界')
# 图例
plt.legend()
plt.show()
正余弦图
- 柱状图
# 切片
# print(string.ascii_uppercase[0: 6])
# 结果：ABCDEF
# 柱状图
from matplotlib import pyplot as plt
# 处理中文乱码
plt.rcParams["font.sans-serif"] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
x = ['口红{}'.format() for x in string.ascii_uppercase[:5]]
y = [randint(200, 500) for _ in range(5)]
print(x)
print(y)
plt.xlabel('口红品牌')
plt.ylabel('价格（元）')
plt.bar(x, y)
plt.show()
柱状图
- 饼图
# 饼图
from matplotlib import pyplot as plt
# 处理中文乱码
plt.rcParams["font.sans-serif"] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
from random import randint
counts = [randint(3500, 9000) for _ in range(6)]
labels = ['员工{}'.format(x) for x in string.ascii_lowercase[:6]]
# 距离圆心点距离
explode = [0.1, 0, 0, 0, 0, 0]
color = ['red', 'purple', 'blue', 'yellow', 'gray', 'green']
plt.pie(counts, explode=explode, shadow=True, labels=labels, autopct='%1.1f%%', colors=color)
plt.axis('equal')
plt.legend(loc=2)
plt.show()
饼图
- 散点图
from matplotlib import pyplot as plt
import numpy as np
# 均值为 0 标准差为 1 的正态分布数据
x = np.random.normal(0, 1, 100)
y = np.random.normal(0, 1, 100)
plt.scatter(x, y)
plt.show()
散点图
- 有透明度的散点图
from matplotlib import pyplot as plt
import numpy as np
x = np.random.normal(0, 1, 100)
y = np.random.normal(0, 1, 100)
# alpha透明度
plt.scatter(x, y, alpha=0.1)
plt.show()
有透明度的散点图
5、绘制 三国top10人物 饼图
with open('./novel/threekingdom.txt', 'r', encoding='utf-8') as f:
words = f.read()
counts = {} # counts = {'姓名':出现频率}
excludes = {"将军", "却说", "丞相", "二人", "不可", "荆州", "不能", "如此", "商议",
"如何", "主公", "军士", "军马", "左右", "次日", "引兵", "大喜", "天下",
"东吴", "于是", "今日", "不敢", "魏兵", "陛下", "都督", "人马", "不知",
"孔明曰", "玄德曰", "刘备", "云长"}
# 2.分词
words_list = jieba.lcut(words)
print(words_list)
for word in words_list:
if len(word) <= 1:
continue
else:
# 字典.get(k) 如果字典中没有这个键 ，(返回NONE)添加一个默认值：0
counts[word] = counts.get(word, 0) + 1
print(counts)
# 3.词语过滤，删除无关词，重复词
counts['孔明'] = counts['孔明'] + counts['孔明曰']
counts['玄德'] = counts['玄德'] + counts['玄德曰'] + counts['刘备']
counts['关公'] = counts['关公'] + counts['云长']
for word in excludes:
del counts[word]
# 4.排序[(), ()]
items = list(counts.items())
# print(items)
# 普通排序
# def sort_by_count(x):
# return x[1]
# items.sort(key=sort_by_count, reverse=True)
# 用列表解析排序
items.sort(key=lambda x: x[1], reverse=True)
# print(items)
li = [] # ['孔明', '孔明', '孔明',..., '曹操'...]
counts = [] # 人物名
labels = [] # 次数
for i in range(10):
# 序列解包
role, count = items[i]
counts.append(count)
labels.append(role)
# 距离圆心点距离
explode = [0.1, 0, 0, 0, 0, 0, 0, 0, 0, 0]
color = ['red', 'purple', 'blue', 'yellow', 'gray', 'green', 'pink']
plt.pie(counts, explode=explode, shadow=True, labels=labels, autopct='%1.1f%%')
plt.axis('equal')
plt.legend(loc=2)
plt.show()
三国Top10人物饼图
6、红楼梦 Top10 人物分析
import jieba
from wordcloud import WordCloud
import imageio
# 1.读取小说内容
with open('./novel/all.txt', 'r', encoding='utf-8') as f:
words = f.read()
counts = {} # counts = {'姓名':出现频率}
excludes = {"什么", "一个", "我们", "你们", "如今", "说道", "太太", "知道", "姑娘",
"起来", "这里", "出来", "众人", "那里", "自己", "一面", "只见", "两个",
"没有", "怎么", "不是", "不知", "这个", "听见", "奶奶", "老太太", "不知",
"这样", "进来", "咱们", "就是", "东西", "告诉", "回来", "只是", "大家",
"老爷", "只得", "丫头", "这些", "他们", "不敢", "出去", "所以", "贾宝玉",
"林黛玉", "薛宝钗", "凤姐儿", "王熙凤"}
# 2.分词
words_list = jieba.lcut(words)
print(words_list)
for word in words_list:
if len(word) <= 1:
continue
else:
# 更新字典中的值
# counts[word] = 取出字典中原来键对应的值 ＋ 1
# counts[word] = counts[word] + 1
# 字典.get(k) 如果字典中没有这个键 ，(返回NONE)添加一个默认值：0
counts[word] = counts.get(word, 0) + 1
print(counts)
# 3.词语过滤，删除无关词，重复词
counts['宝玉'] = counts['宝玉'] + counts['贾宝玉']
counts['黛玉'] = counts['黛玉'] + counts['林黛玉']
counts['宝钗'] = counts['宝钗'] + counts['薛宝钗']
counts['贾母'] = counts['贾母'] + counts['老太太']
counts['凤姐'] = counts['凤姐'] + counts['凤姐儿'] + counts['王熙凤']
for word in excludes:
del counts[word]
# 4.排序[(), ()]
items = list(counts.items())
print(items)
# def sort_by_count(x):
# return x[1]
# items.sort(key=sort_by_count, reverse=True)
# 用列表解析排序
items.sort(key=lambda x: x[1], reverse=True)
# print(items)
li = [] # ['孔明', '孔明', '孔明',..., '曹操'...]
for i in range(10):
# 序列解包
role, count = items[i]
print(role, count)
# _ 是告诉看代码的人，循环里面不需要使用临时变量
for _ in range(count):
li.append(role)
# 得出结论
# 绘制中文词云，在WordCloud()里面设置参数
text = ' '.join(li)
wc = WordCloud(
font_path='msyh.ttc',
background_color='white',
width=800,
height=600,
# 相邻两个重复词之间的匹配，关掉
collocations=False
).generate(text)
wc.to_file('红楼梦TOP10人物词云.png')
红楼梦TOP10人物词云