pd.read_excel()練習

# -*- coding: utf-8 -*-
import pandas as pd
import numpy as np

df11=pd.read_excel("C:/Users/Administrator/Desktop/20190527/表1.xlsx")
df12=pd.read_excel("C:/Users/Administrator/Desktop/20190527/表2.xlsx")
df13=pd.read_excel("C:/Users/Administrator/Desktop/20190527/表3.xlsx")
df11['LipidIon.1']=df11['LipidIon.1'].apply(lambda x : x.strip())
df12['LipidIon.1']=df12['LipidIon.1'].apply(lambda x : x.strip())
df13['LipidIon.1']=df13['LipidIon.1'].apply(lambda x : x.strip())
df11['LipidIon']=df11['LipidIon'].apply(lambda x : x.strip())
df12['LipidIon']=df12['LipidIon'].apply(lambda x : x.strip())
df13['LipidIon']=df13['LipidIon'].apply(lambda x : x.strip())

df11['category']=df11['LipidIon'].apply(lambda x : x.split("+")[-1])
df12['category']=df12['LipidIon'].apply(lambda x : x.split("+")[-1])
df13['category']=df13['LipidIon'].apply(lambda x : x.split("+")[-1])
str(df11['LipidIon.1'])

df1=df11[['LipidIon','LipidIon.1','category','Area']]
df2=df12[['LipidIon','LipidIon.1','category','Area']]
df3=df13[['LipidIon','LipidIon.1','category','Area']]
df=pd.concat([df1,df2,df3],axis=0)
df=df.sort_values(by=['LipidIon'],inplace=False, kind='quicksort', na_position='last') 
df=df.reset_index(drop=True)    #重置索引

#1.功能1:過濾信息並彙總表4
count=0
dfdf=df.groupby(['LipidIon'], as_index = False)
data=pd.DataFrame(columns=['LipidIon','category','LipidIon.1','Area_i','Area_sum'])

#1.先對數據做分羣處理
class_name = list(dfdf[['LipidIon','category']])

#for line in class_name[62:63]:
for line in class_name:
    #對每一個分組內部,統計三類子分組中的最大值: max(max(每一個分組))
    group1_name=line[0]
    group1_data=line[1].groupby(['category'], as_index = False)
    class_name1 = list(group1_data[['LipidIon.1','category']])
    #統計每個分組中的最大值,並計算總的峯值面積
    #篩選出來最大值所對應的行
    max1_row=class_name1[0][1][class_name1[0][1].Area==class_name1[0][1]['Area'].max()]
    #計算最大值所在那個分組的所有峯值面積和
    #max_key=max1_row['category']
    area0=0
    class_name1[0][1].index=range(len(class_name1[0][1]))
    area0=class_name1[0][1]['Area'].sum()
    #重新組織數據存表
    #type(max1_row['LipidIon'])    //Series
    data.loc[count,'LipidIon']=max1_row['LipidIon'].values[0]
    data.loc[count,'LipidIon.1']= max1_row['LipidIon.1'].values[0]
    data.loc[count,'category']= max1_row['category'].values[0] 
    data.loc[count,'Area_i']= max1_row['Area'].values[0] 
    data.loc[count,'Area_sum']=area0
    count+=1
data['Area_i']= data['Area_i'].apply(lambda x:np.float(x))  
data['Area_sum']= data['Area_sum'].apply(lambda x:np.float(x))  
data=data.sort_values(by=['LipidIon'],inplace=False, kind='quicksort', na_position='last') 
data=data.reset_index(drop=True)    #重置索引
data.to_excel("C:/Users/Administrator/Desktop/20190527/表4_去重.xlsx")


#讀入表4
biao4=pd.read_excel("C:/Users/Administrator/Desktop/20190527/表4_去重.xlsx")
biao4_name=sorted(set(biao4['LipidIon.1']))

#2.功能2:從表4中刪除掉只出現一次的行
freq_dict=dict()
for i in range(len(biao4_name)):
    freq_dict[biao4_name[i]]=0


for i in range(data.shape[0]):
    freq_dict[data.loc[i,'LipidIon.1']]+=1
#篩選出來大於2的行    
lengt2=[]
for key,value in freq_dict.items():
    if value>1:
        lengt2.append(key)
dfdf33=pd.DataFrame(lengt2,columns=['LipidIon.1'])
ss_df=pd.merge(dfdf33,data)   #做一次數據關聯,關聯出來出現行數大於2的行。
ss_df.to_excel("C:/Users/Administrator/Desktop/20190527/表4.xlsx")



#讀入總表
sum_df=pd.read_excel("C:/Users/Administrator/Desktop/20190527/總表.xlsx")
sum_df.columns.tolist()
sum_name=sorted(set(sum_df['LipidMolec']))
sum_df=sum_df[['LipidMolec','MainMScore[c]','NMArea[c]']]
sum_df.columns=['LipidIon.1','MainMScore[c]','NMArea[c]']

#讀入表4
biao4=pd.read_excel("C:/Users/Administrator/Desktop/20190527/表4.xlsx")
biao4_name=sorted(set(biao4['LipidIon.1']))

#求交集=>表5
interSet=sorted(set(sum_name) & set(biao4_name))
biao41=biao4[biao4['LipidIon.1'].apply(lambda x : x in interSet )]
biao41.to_excel("C:/Users/Administrator/Desktop/20190527/表5.xlsx")

#求表4沒有但總表有的=>表6
onlySumhas=sorted(set(sum_name) - set(biao4_name))
biao42=sum_df[sum_df['LipidIon.1'].apply(lambda x : x in onlySumhas )]
biao42.to_excel("C:/Users/Administrator/Desktop/20190527/表6.xlsx")

#表4有但總表沒有的=>表7
onlybiao4has=sorted(set(biao4_name) - set(sum_name))
biao43=biao4[biao4['LipidIon.1'].apply(lambda x : x in onlybiao4has )]
biao43.to_excel("C:/Users/Administrator/Desktop/20190527/表7.xlsx")