Question

我有3个pandas数据框，每个都有不同的行数和一些相似的列，我需要将所有数据合并在一起

mydata = [0]*3
dataA = {'First':  [500],'Second': ['Sone']}
mydata[0] = pd.DataFrame(dataA,columns=['First','Second'])
dataB = {'First':  [500,500],'Third': [0.5,0.6]}
mydata[1] = pd.DataFrame(dataB,columns=['First','Third'])
dataC = {'First':  [500,500,500],'Fourth': ['Fone', 'Ftwo','Fthree'],'Fifth': [23, 24, 25]}
mydata[2] = pd.DataFrame(dataC,columns=['First','Fourth','Fifth'])

合并数据看起来像

merge_data = {'First':  [500,500,500,500,500,500],'Second': ['Sone','Sone','Sone','Sone','Sone','Sone'],'Third': [0.5,0.6,0.5,0.6,0.5,0.6],'Fourth': ['Fone', 'Fone', 'Ftwo', 'Ftwo', 'Fthree','Fthree'],'Fifth': [23, 23, 24, 24, 25, 25]}
merge_df = pd.DataFrame(merge_data,columns=['First','Second','Third','Fourth','Fifth'])

数据附加产生Nan行

merge_data = mydata[0].copy()
for i in np.arange(1, len(mydata)):
    merge_data = merge_data.append(mydata[i], sort=False)

并合并丢失的行

merge_data = pd.merge(mydata[0], mydata[1], left_index=True, right_index=True)

是否可以合并为merged_df

Answer 1

您必须在'First'列上进行合并：

pd.merge(mydata[0], mydata[1], on='First').merge(mydata[2], on='First')

获得：

   First Second  Third  Fourth  Fifth
0    500   Sone    0.5    Fone     23
1    500   Sone    0.5    Ftwo     24
2    500   Sone    0.5  Fthree     25
3    500   Sone    0.6    Fone     23
4    500   Sone    0.6    Ftwo     24
5    500   Sone    0.6  Fthree     25

仅Fourth和Fifth列仍在此处对齐，而merge_df数据帧中没有...

合并熊猫数据框与不同的行？

1 个答案: