数据分析

matplotlib模块

'''from matplotlib import pyplot as plt x=range(0,20,2) y=[12,1,3,5,6,56,62,13,35,66] #设置图片大小，dpi设置清晰度 plt.figure(figsize=(20,10),dpi=80) #绘图 plt.plot(x,y) #保存 #plt.savefig("./1.jpg") #设置x轴的刻度 plt.xticks(range(0,20)) plt.yticks(range(min(y),max(y)+1)) #展示图形 plt.show()''' '''from matplotlib import pyplot as plt import random plt.rcParams['font.sans-serif'] = [u'SimHei'] plt.rcParams['axes.unicode_minus'] = False x=range(120) y1=[random.randint(20,36) for i in range(120)] y2=[random.randint(20,36) for i in range(120)] plt.figure(figsize=(15,8),dpi=80) #label添加折线图名称，color颜色，要跟上legend才能显示 plt.plot(x,y1,label="jackson",color="pink") plt.plot(x,y2,label="杜",color="purple") #图形注释 plt.legend() #调整x轴的刻度 z=list(x)[::10] labels=["hello {}".format(i) for i in z] plt.xticks(z,labels,rotation=45)#rotation旋转度数 plt.xlabel("时间") plt.ylabel("温度 单位（℃）") plt.title("不同时间的温度变化") #绘制网格 plt.grid() plt.show()''' '''#绘制温度散点图 from matplotlib import pyplot as plt import random plt.rcParams['font.sans-serif'] = [u'SimHei'] plt.rcParams['axes.unicode_minus'] = False y_3=[random.randint(12,20) for i in range(31)] y_10=[random.randint(20,30) for i in range(31)] x_3=range(1,32) x_10=range(51,82) #使用scatter绘制散点图 plt.scatter(x_3,y_3,label="三月份") plt.scatter(x_10,y_10,label="十月份") #添加图例 plt.legend(loc="upper left") #调整x轴的刻度 _x=list(x_3)+list(x_10) _xlabels=["三月{}日".format(i) for i in x_3]+["十月{}日".format(i-50) for i in x_10] plt.xticks(_x[::3],_xlabels[::3],rotation=45) #添加描述信息 plt.xlabel("时间") plt.ylabel("温度") plt.title("每天温度变化") plt.show()''' '''from matplotlib import pyplot as plt plt.rcParams['font.sans-serif'] = [u'SimHei'] plt.rcParams['axes.unicode_minus'] = False a=["星际穿越","1917","误杀","急速车王","哪吒之魔童降世","巨齿鲨","大鱼海棠","大话西游之大圣娶亲","风声","美人鱼"] b=[20,19,15,16,24,25,18,16,11,21] plt.figure(figsize=(15,8),dpi=80) #绘制条形图，barh表示横向 plt.barh(range(len(a)),b,height=0.3,color="pink") plt.yticks(range(len(a)),a) plt.show()'''

numpy模块

'''import numpy as np import numpy.matlib as matlib arr=np.array([[88,0,11,0,17,0,46,0,7,0,75,0], [28,0,33,0,84,0,96,0,88,0,44,0], [5,0,4,0,71,0,88,0,88,0,50,0], [54,0,34,0,15,0,77,0,88,0,15,0], [6,0,85,0,22,0,11,0,12,0,92,0]]) print(arr) print('\n') print(arr[1])#取一行 print('\n') print(arr[:,0:2])#切片出前两列 print('\n') print(arr[0:2,0:2])#取出前两行的前两列 print('\n') print(arr[::-1,::-1])#行列反转 print('\n') print(arr.reshape(60,))#多维数组变成一维数组 print('\n') print(arr.reshape(12,-1))#-1自动计算行或者列 print('\n') s1=np.array([[1,2,3],[4,5,6]]) s2=np.array([[1,3,5],[2,4,6]]) M=np.concatenate((s1,s2),axis=0)#1表示横向连接，0表示纵向链接 print(M) print('\n') s3=np.array([5.6,6.6,7.6]) print(np.around(s3,0))#对小数进行四舍五入 print('\n') s4=np.random.randint(50,100,size=(3,5)) print(s4,'\n') print(np.amin(s4,axis=0))#找出一列中的最小值 print(np.ptp(s4,axis=0))#最小值与最大值之间的差 print(np.median(s4,axis=0))#找出一组数据的中间值 print(np.std(s3),np.var(s3))#求标准差和方差 print('\n') N=matlib.ones(shape=(2,3)) print(N) print('\n') k=matlib.eye(n=5,M=5,k=0)#n表示行数，M表示列数，k表示对角线索引 print(k,'\n')#转置为arr.T print(matlib.identity(5)) print('\n') a1=([[1,2],[3,4]]) b1=([[2,2],[5,1]]) print(np.dot(a1,b1))#矩阵相乘''' import numpy as np a1=np.arange(10)#生成0-9存入列表 print(a1) a2=np.array([[1,2,3],[4,5,6]]) print(a2.shape)#shape查看数组的形状 print(a2.reshape(3,2))#reshape修改数组的形状 print(a2.flatten())#展开成一维数组 print(a2.transpose())#transpose、T、swapaxes（0，1）都可以实现转置 print(a2[[0,1],[1,2]])#取多个数字，（0，1），（1，2）为坐标 a3=np.arange(24).reshape(4,6) print(np.where(a3<10,66,99))#把小于10的数换成66，否则为99 a4=np.arange(12).reshape((3,4)).astype("float")#astype转化类型后才能变为nan a4[1,2:]=np.nan print(a4) a5 = np.array([1, 2, 3]) a6 = np.array([4, 5, 6]) res1 = np.vstack((a5, a6))#vstack竖直拼接，hstack水平拼接 print(res1)

pandas模块

import pandas as pd import numpy as np from pandas import Series,DataFrame a=Series(data=[1,2,3,4,5,6]) print(a) dic={ 'a':1, 'b':2, 'c':3 } b=Series(data=dic) print(b) c=Series(data=[1,2,3],index=['数学','语文','英语']) print(c) print(c.head(2))#显示前两条数据 print(c.tail(2))#后两条 d=Series(data=[1,2,6,4,3,4,2,3,5,9,7,9]) print(d.unique())#去重 s1=Series(data=[1,2,3]) s2=Series(data=[4,5,6]) print(s1+s2) s3=Series(data=[1,2,3,4,5,6],index=['a','b','c','d','e','f']) s4=Series(data=(2,3,6),index=['c','e','f']) s=s3+s4 e=s[s.notnull()]#去除为null的值 print(e) df=DataFrame(data=np.random.randint(0,100,size=(5,6))) print(df) dic={ 'name':['jackosn','jay','jaky'], 'salary':[20000,10000,15000] } df2=DataFrame(data=dic,index=[1,2,3]) print(df2) #print(df2.loc[1])#取出一行数据，或者使用df2.iloc[0] #print(df2[0:2])#切片前两行 #print(df2.iloc[:,0:1])#切列 f=df2.drop(labels='salary',axis=1)#axis=0表示行，=1表示列 print(f) import numpy as np import pandas as pd from pandas import DataFrame '''df=DataFrame(np.random.randint(0,10,size=(5,6))) #下面两种替换方式 df.replace(to_replace=0,value='zero') print(df.replace(to_replace={0:'abc',5:13})) #将指定列中的替换 df.iloc[2]=[0,0,0,0,0,0] print(df.replace(to_replace={3:0},value=666)) dic={'name':['张三','李四','王五'], 'salary':[20000,15000,10000]} s=DataFrame(data=dic) print(s) dic={ '张三':'tom', '李四':'jay', '王五':'jerry' } #映射关系表 print(s['name'].map(dic)) #计算税后收入，超过3000收50%的税 def after_sal(s): return s-(s-3000)*0.5 s['after_sal']=s['salary'].map(after_sal) print(s) #apply操作： #可以将行或列数据进行运算 def fun(s): s=s.sum() print(s) s.apply(fun)''' '''df=DataFrame({ 'color':['red','purple','blue','green','pink'], 'value':np.random.randint(10,size=5) }) print(df) #映射df中的数据 new_index={0:'zero',1:'one',2:'two',3:'three',4:'four'} #映射列索引 new_col={'color':'cc','value':'vv'} print(df.rename(index=new_index,columns=new_col))''' '''#排序实现的随机抽样 df=DataFrame(np.random.randint(100,size=(20,3)),columns=['A','B','C']) print(df.take(np.random.permutation(3),axis=1).take(np.random.permutation(20),axis=0)) ''' #数据的分类处理 '''df=DataFrame({'item':['a','b','c','b','c','a'], 'price':[1,6,8,5,11,3], 'color':['pink','purple','blue','blue','pink','yellow'], 'weight':[16,12,18,20,12,30]}) average=df.groupby(by='item').mean()['price'] dic=average.to_dict() df['average']=df['item'].map(dic) print(df)''' data=pd.read_csv('C:/Users/Administrator/Desktop/1.txt') print(data) import pandas as pd import numpy as np dates = pd.date_range('20130101', periods=6) df = pd.DataFrame(np.random.randn(6, 4), index=dates, columns=list('ABCD')) #输出日期格式，六行四列 print(df) df2 = pd.DataFrame({'A': 1., 'B': pd.Timestamp('20200406'), 'C': pd.Series(1, index=list(range(4)), dtype='float32'), 'D': np.array([3] * 4, dtype='int32'), 'E': pd.Categorical(["test", "train", "test", "train"]), 'F': 'foo'}) print(df2) df3 = pd.DataFrame(np.random.randn(10, 4)) print(df3) #分成三组 pieces = [df3[:3], df3[3:7],df3[7:]] print(pieces) #合并所有组 print(pd.concat(pieces)) import pandas as pd left = pd.DataFrame({'key': ['foo', 'foo'], 'lval': [1, 2]}) right = pd.DataFrame({'key': ['foo', 'foo'], 'rval': [4, 5]}) M=pd.merge(left, right, on='key') print(left) print(right) #连接 print(M) import pandas as pd import numpy as np df = pd.DataFrame(np.random.randn(8, 4), columns=['A', 'B', 'C', 'D']) s = df.iloc[3] #追加3这一行的数在末尾 df.append(s, ignore_index=True) import pandas as pd import numpy as np tuples = list(zip(*[['bar', 'bar', 'baz', 'baz', 'foo', 'foo', 'qux', 'qux'], ['one', 'two', 'one', 'two', 'one', 'two', 'one', 'two']])) #zip(*)表示取各个list的第x项，作为返回的二维数组的第x项中的一维数组中的元素 index = pd.MultiIndex.from_tuples(tuples, names=['first', 'second']) #MultiIndex表示多级索引，它是从Index继承过来的，其中多级标签用元组对象来表示 df = pd.DataFrame(np.random.randn(8, 2), index=index, columns=['A', 'B']) stacked = df.stack() #stack() 方法把 DataFrame 列压缩至一层 print(stacked)