python数据分析 生成词云图、jieba分词 从最简单的数据集入门

 1.仅仅使用了一列文字数据进行生成词云图，我发现生成的图很丑，就当做是个简单的入门吧！哈哈

首先取了一列数据，并且转化为txt 文件

#词云图尝试
import numpy as np
import pandas as pd
from wordcloud import WordCloud
import PIL.Image as image
import csv

"""
没用到的东西
wind_sky = pd.read_csv('datadatadata1.csv',usecols = [5],header=None)
print(wind_sky)
print(type(wind_sky))

"""

#取数据
wind_sky = pd.read_csv('datadatadata1.csv',usecols = [5],header=None)
print(wind_sky)
print(type(wind_sky))

wind_sky_list = []
filename = 'datadatadata1.csv'
with open(filename,'r', encoding='UTF-8') as f:
    reader = csv.reader(f)
    header_row = next(reader)
    column1 = [row[5] for row in reader]
    wind_sky_list.append(column1)
print(wind_sky_list)

#转为txt文件（主要是自己不太会，所以就先这样子转一下）
def text_save(filename, data):  # filename为写入txt文件的路径，data为要写入数据列表.
    file = open(filename, 'a')
    for i in range(len(data)):
        s = str(data[i]).replace('[', '').replace(']', '')  # 去除[],这两行按数据不同，可以选择
        s = s.replace("'", '').replace(',', '') + '\n'  # 去除单引号，逗号，每行末尾追加换行符
        file.write(s)
    file.close()
text_save('wind_sky_list.txt',wind_sky_list)

 

#正式开始画词云图
from wordcloud import WordCloud
import PIL.Image as image
import jieba
import numpy as np
def trans_cn(text):
    word_list = jieba.cut(text)
    result = " ".join(word_list)
    return result;

with open("wind_sky_list5.txt") as fp:
    text = fp.read()
    text = trans_cn(text)
    print(text)
    #选了一张背景图片
    mask = np.array(image.open("ciyun.png"))

    WordCloud = WordCloud(
        #背景图片
        mask = mask,
        #如果txt里面有中文的话一定要加这一行代码，不然出来的图会是空的框框！！！
        font_path = "C:\\Windows\\Fonts\\msyh.ttc"
    ).generate(text)
    image_produce = WordCloud.to_image()
    image_produce.show()

上面的图画完之后，感觉词太少了导致非常丑，所以现在我又加了一些数据，也就是又从我的数据集中多取了列数据。这里取数据的时候还发生了一些小插曲，就是我要取得那一列数据里面有波浪号，所以要去一下，相当于停用词吧，我知道有那更简单的方式，不过作为刚刚入门的小白，我这边还是写的比较啰嗦。

2.再次尝试画词云图（多加了一些数据）

2.1去多余符号

import csv
import numpy
list_no_bolang = []
import  pandas as pd
list_no_bolang2 = []
with open('fy_tianqi_2019-2020.csv', 'rt',encoding='UTF-8') as f:
    reader = csv.reader(f)
    header_row = next(reader)
    for row in reader:
        for word in row:
            str2 = word.replace('~', '')
            list_no_bolang.append(str2)

for row1 in list_no_bolang:
    #print(row1)
    str3 = row1.replace('℃', '')
    list_no_bolang2.append(str3)

print(list_no_bolang)
print(list_no_bolang2)
print(numpy.shape(list_no_bolang))
#print(numpy.shape(list_no_bolang2))


list_no_bolang_np=numpy.array(list_no_bolang2).reshape(345,9)
name = ["ymd","bWendu","yWendu","tianqi","fengxiang","fengli","aqi","aqiInfo","aqiLevel"]
csv_list = pd.DataFrame(columns=name, data=list_no_bolang_np)
csv_list.to_csv('for new ciyun//sky4.csv')

2.2取数据，转化为TXT文件

import numpy as np
import pandas as pd

import csv


#取4,5两列
wind_sky_list = []
filename = 'sky4.csv'
with open(filename,'r', encoding='UTF-8') as f:
    reader = csv.reader(f)
    header_row = next(reader)

    column1 = [row[4] for row in reader]
    wind_sky_list.append(column1)
    print(np.shape(wind_sky_list))
with open(filename,'r', encoding='UTF-8') as f:
    reader = csv.reader(f)
    header_row = next(reader)
    column2 = [row[5] for row in reader]
    wind_sky_list.append(column2)
    print(np.shape(wind_sky_list))


print(wind_sky_list)




#转为txt文件（主要是自己不太会，所以就先这样子转一下）
def text_save(filename, data):  # filename为写入txt文件的路径，data为要写入数据列表.
    file = open(filename, 'a')
    for i in range(len(data)):
        s = str(data[i]).replace('[', '').replace(']', '')  # 去除[],这两行按数据不同，可以选择
        s = s.replace("'", '').replace(',', '') + '\n'  # 去除单引号，逗号，每行末尾追加换行符
        file.write(s)
    file.close()
text_save('wind_sky_list5.txt',wind_sky_list)

2.3开始jieba分词并且画词云图

#正式开始画词云图
from wordcloud import WordCloud
import PIL.Image as image
import jieba
import numpy as np
def trans_cn(text):
    word_list = jieba.cut(text)
    result = " ".join(word_list)
    return result;

with open("wind_sky_list5.txt") as fp:
    text = fp.read()
    text = trans_cn(text)
    print(text)
    #选了一张背景图片
    mask = np.array(image.open("ciyun.png"))

    WordCloud = WordCloud(
        #背景图片
        mask = mask,
        #如果txt里面有中文的话一定要加这一行代码，不然出来的图会是空的框框！！！
        font_path = "C:\\Windows\\Fonts\\msyh.ttc"
    ).generate(text)
    image_produce = WordCloud.to_image()
    image_produce.show()

 分词后的示例：

 画出来的词云图：（还是有点丑 就是先练习练习：）：两次运行结果不太一样：不过看起来字号大小是一样的