```python import pandas as pd import numpy as np df = pd.read_excel('D://test/ppp.xlsx',sheet_name="Phylum") # print(df) # df1=df.groupby(by='Phylum') # # df # df1.to_excel("D://test//PP_Des.xlsx") # df['X.mean'] = df['SUM'].mean() #df.groupby(by='Phylum').describe() # # df1 = df.query('Phylum=="p__Firmicutes"') # # df1['xi-x'] =np.power(df1['SUM'] - df['X.mean'],2) # std = np.power(df1['xi-x'].sum()/df['Phylum'].count(),0.5) # print(std) # df['S'] = np.power(df['SUM']-df['SUM'].mean(),2) # # df['x'] = np.power(df["S"],0.5)/df['Phylum'].count()-1 # print(df['Phylum'].count()) # # # # print(df) # df1 = df.groupby(by='Phylum').describe() # df1 = df.groupby(by='Family').describe() # df1 = df.describe() df2 = df.groupby (by='Phylum').sum ( ) df2.index # print(df2.index) # # print(df2.index) name = df2.index # # print(name) # # print(name) list1=[] # print(df['Phylum']) my = {} x=pd.Dataframe() for i in name: # print(i) df1 = df.query('Phylum=="%s"'%str(i)) count = df1["Phylum"].count() print(df1) df1['per'] = df1['SUM'] / df1['SUM'].sum ( ) df1['X.mean'] = df1['per'].mean() df1['(xi-x)^2'] =np.power(df1['per'] - df1['X.mean'],2) H = df1['(xi-x)^2'].sum() x[i] = H df1 = np.power(H/(df1['Phylum'].count()-1),0.5) # count = df1['Phylum'].count( ) xh = pd.Dataframe({'Phylum': i, 'std': H, 'count': count},index=[0]) list1.append(xh) df4=pd.concat(list1) df4.to_excel('D://test/xh1.xlsx') print(xh) # # df3 = pd.Dataframe.from_dict(my) # # df3 = pd.Dataframe(my) # # # df3 = pd.Dataframe.from_dict(my,orient = 'index') # print(df4) # print(list1) # df3.to_excel ('D://test/xh1.xlsx') # # # # print(df4) # df3 = df1.groupby (by='Phylum').describe ( ) # list1.append(df3) # df4=pd.concat(list1) # df4["all.sum"] = df2['SUM'].sum() # # # df4.to_excel ('D://test/xh1.xlsx') # print(df4) # # df4= pd.Dataframe(list1) # # print(df4) # # df1 = df.query('Phylum=="p__Firmicutes"') # # # # df1['per'] = df1['SUM'] /df1['SUM'].sum() # # # # # # df2 = df1.groupby(by='Phylum').describe() # # # # # # # # # df2 = df.groupby(by='Phylum').describe() # # # # # df1['perc'] = df1['SUM']/df1['SUM'].sum() # # # # df2.to_excel('D://test/xh2.xlsx') # df1.to_excel('D://test/xxh.xlsx')
欢迎分享,转载请注明来源:内存溢出
评论列表(0条)