Python 是否将已排序的表从数据帧发送到前端？

我正在对财务数据进行一些计算，我的原始json如下所示：

[
  {
    "company": "Apple Inc",
    "earnings_yield": "0.032220032460810885",
    "roic": "0.18313585674458502"
  },
  {
    "company": "Microsoft Corporation",
    "earnings_yield": "0.0360941393518613",
    "roic": "0.1955875358727057"
  }
]

{
  "company": {
    "0": "Apple Inc",
    "1": "Microsoft Corporation"
  },
  "earnings_yield": {
    "0": 0.0322200325,
    "1": 0.0360941394
  },
  "roic": {
    "0": 0.1831358567,
    "1": 0.1955875359
  },
  "roic_rank": {
    "0": 1,
    "1": 2
  },
  "ey_rank": {
    "0": 1,
    "1": 2
  },
  "sum_rank": {
    "0": 2,
    "1": 4
  }
}

'{"0":{"company":{"0":"Apple Inc"},"earnings_yield":{"0":"0.032220032460810885"},"roic":{"0":"0.20313585674458502"},"roic_rank":{"0":2.0},"ey_rank":{"0":1.0},"sum_rank":{"0":3.0}},"1":{"company":{"1":"Microsoft Corporation"},"earnings_yield":{"1":"0.0360941393518613"},"roic":{"1":"0.1955875358727057"},"roic_rank":{"1":1.0},"ey_rank":{"1":2.0},"sum_rank":{"1":3.0}}}'

我通过pandas数据框运行json，并在其中对它们进行排序：

df = pd.DataFrame(companies)
df['roic_rank'] = df['roic'].rank()
df['ey_rank'] = df['earnings_yield'].rank()
df['sum_rank'] = df['ey_rank'] + df['roic_rank'].rank()

返回：

                 company  earnings_yield    roic  roic_rank  ey_rank  sum_rank
0              Apple Inc          0.0322  0.1831        1.0      1.0       2.0
1  Microsoft Corporation          0.0361  0.1956        2.0      2.0       4.0

将已排序的表返回到前端，以便以升序/降序（在

sum\u rank

上）的下拉列表形式很好地显示它的好方法是什么

如果我只做

df.to_json（）

，我所有关于熊猫的工作都会被扔出窗口，因为json是这样返回的：

[
  {
    "company": "Apple Inc",
    "earnings_yield": "0.032220032460810885",
    "roic": "0.18313585674458502"
  },
  {
    "company": "Microsoft Corporation",
    "earnings_yield": "0.0360941393518613",
    "roic": "0.1955875358727057"
  }
]

{
  "company": {
    "0": "Apple Inc",
    "1": "Microsoft Corporation"
  },
  "earnings_yield": {
    "0": 0.0322200325,
    "1": 0.0360941394
  },
  "roic": {
    "0": 0.1831358567,
    "1": 0.1955875359
  },
  "roic_rank": {
    "0": 1,
    "1": 2
  },
  "ey_rank": {
    "0": 1,
    "1": 2
  },
  "sum_rank": {
    "0": 2,
    "1": 4
  }
}

'{"0":{"company":{"0":"Apple Inc"},"earnings_yield":{"0":"0.032220032460810885"},"roic":{"0":"0.20313585674458502"},"roic_rank":{"0":2.0},"ey_rank":{"0":1.0},"sum_rank":{"0":3.0}},"1":{"company":{"1":"Microsoft Corporation"},"earnings_yield":{"1":"0.0360941393518613"},"roic":{"1":"0.1955875358727057"},"roic_rank":{"1":1.0},"ey_rank":{"1":2.0},"sum_rank":{"1":3.0}}}'

你可以试试groupby

json = df.groupby(level=0).apply(lambda x: x.to_dict(orient='records'))
json = json.to_json()

输出将如下所示：

[
  {
    "company": "Apple Inc",
    "earnings_yield": "0.032220032460810885",
    "roic": "0.18313585674458502"
  },
  {
    "company": "Microsoft Corporation",
    "earnings_yield": "0.0360941393518613",
    "roic": "0.1955875358727057"
  }
]

{
  "company": {
    "0": "Apple Inc",
    "1": "Microsoft Corporation"
  },
  "earnings_yield": {
    "0": 0.0322200325,
    "1": 0.0360941394
  },
  "roic": {
    "0": 0.1831358567,
    "1": 0.1955875359
  },
  "roic_rank": {
    "0": 1,
    "1": 2
  },
  "ey_rank": {
    "0": 1,
    "1": 2
  },
  "sum_rank": {
    "0": 2,
    "1": 4
  }
}

'{"0":{"company":{"0":"Apple Inc"},"earnings_yield":{"0":"0.032220032460810885"},"roic":{"0":"0.20313585674458502"},"roic_rank":{"0":2.0},"ey_rank":{"0":1.0},"sum_rank":{"0":3.0}},"1":{"company":{"1":"Microsoft Corporation"},"earnings_yield":{"1":"0.0360941393518613"},"roic":{"1":"0.1955875358727057"},"roic_rank":{"1":1.0},"ey_rank":{"1":2.0},"sum_rank":{"1":3.0}}}'

---

预期的json输出是什么？可能是

df.to_dict（“records”）

？一些带有对象数组的排序json，其中第一个元素是具有最低“sum_rank”的索引。您能解释一下这是如何工作的吗？理想情况下，我希望订单基于sum_秩。您的lambda在表中按索引对其进行排序？因此，即使我运行任何

df=df.sort\u值（按class='sum\u rank'）

或任何东西。索引将保持不变-不确定这是否解决了我的问题。好的，您可以将df的索引设置为sum_rank，然后对其应用groupby

df=df.set_index（'sum_rank'）

json=df.groupby（level=0）。apply（lambda x:x.to_dict（orient='records'））

json=json.to_json（）

有关groupby引用，请参阅（）