python - 如何在 pandas 和 python 中创建汇总数据

Question

我正在尝试格式化来自外部源的一些数据并进行一些排序和分组。以下是我当前正在使用的片段。我正在尝试进行排序和分组。

pmtAccount = []
pmtAmount = []

for payment in payments_sent:
    print ("{} {:>15}".format(payment['to'], payment['amount']))
    pmtAmount.append(payment['amount'])
    pmtAccount.append(payment['to'])


#-- Attempt to summarize output
df = pd.DataFrame(list(zip(pmtAccount, pmtAmount)), columns =['To', 'Amount'])    

result = df.groupby('To')['Amount'].agg(['count','sum']).reset_index()
result['count'] = result['count'].astype(str) + ' Transactions'
print(result)

当前输出:

#-- Output of the line print ("{} {:>15}".format(payment['to'], payment['amount']))

199SJFW1K     100.0000000
ABC123ABC     100.0000000
ABC123ABC      50.0000000
ABC123ABC     300.0000000
123ABC123     200.0000000
123ABC123      50.0000000
123ABC123     100.0000000
ABC123ABC     300.0000000
ABC123ABC     250.0000000
XYXY12XYX     250.0000000
XYXY12XYX     500.0000000
1OPR12DRT      41.0000000
KSJDHW812       0.1900000

#-- 
                                                  To           count                  sum
0  ABC123ABC       ...  5 Transactions  100.0000000 50.0000000 300.0000000 300.0000000250...
1  XYXY12XYX       ...  2 Transactions                             250.0000000500.0000000
2  KSJDHW812       ...  1 Transactions                                          0.1900000
3  1OPR12DRT       ...  1 Transactions                                         41.0000000
4  123ABC123       ...  3 Transactions                   200.000000050.0000000100.0000000
5  199SJFW1K       ...  1 Transactions                                        100.0000000

需要的输出:#--排序和分组

0  ABC123ABC    ...  5 Transactions       1,100.0000000 #-- Sum of (100.0000000, 50.0000000, 300.0000000, 300.0000000, 250.0000000)
1  123ABC123    ...  3 Transactions         350.0000000    
2  XYXY12XYX    ...  2 Transactions         750.0000000
3  1OPR12DRT    ...  1 Transactions          41.0000000
4  199SJFW1K    ...  1 Transactions         100.0000000    
5  KSJDHW812    ...  1 Transactions           0.1900000
Total Transactions: 13     Total Amount:  2,341.19 

Answer 1

你的方法是正确的，但是你思考数据后，你没有将 Amount 转换为 Float 类型。

df['Amount'] = df['Amount'].astype(float)

现在您可以分别计算总交易数和总金额:

result = df.groupby('To')['Amount'].agg(['count','sum']).reset_index()
total_transactions = result['count'].sum()
total_amount = result['sum'].sum()

现在包含这些更改的完整代码块将如下所示:

df = pd.DataFrame(list(zip(pmtAccount, pmtAmount)), columns =['To', 'Amount'])
df['Amount'] = df['Amount'].astype(float)

result = df.groupby('To')['Amount'].agg(['count','sum']).reset_index()
total_transactions = result['count'].sum()
total_amount = result['sum'].sum()
result['count'] = result['count'].astype(str) + ' Transactions'

result = result.sort_values(by=['sum', 'count'], ascending=[False, False])
print(result)
print("Total Transactions: {} Total Amount: {}".format(total_transactions, total_amount))

输出:

           To           count      sum
3  ABC123ABC  5 Transactions  1000.00
5  XYXY12XYX  2 Transactions   750.00
0  123ABC123  3 Transactions   350.00
1  199SJFW1K  1 Transactions   100.00
2  1OPR12DRT  1 Transactions    41.00
4  KSJDHW812  1 Transactions     0.19
Total Transactions: 13 Total Amount: 2241.19