从 pandas 中的多索引数据框中选择特定行

Question

I would like to select the last row from a multiindex dataframe and append to a dict of buy and sell signals. For example, given the multiindex dataframe below:

enter image description here

I would like to select the last row indexed (HK.00700 and 2022-06-28 10:39:00), and add to the dict as follows while keeping the last row's multiindices:

enter image description here

The indices in the second pic are slightly different, but the idea is the same.

Answer 1

Reproduce your data

level = [['HK.00700'],[pd.Timestamp('2022-06-28 10:38:00'),pd.Timestamp('2022-06-28 10:39:00')]]
level_index = pd.MultiIndex.from_product(level, names=['code','time_key'])
transaction = {
    'open':[360.6, 360.8],
    'close':[360.6, 361.4],
    'high':[360.8, 361.4],
    'low':[360.4, 360.4],
    'volume':[72500, 116300],
    'upper_band':[360.906089, 361.180835],
    'lower_band':[357.873911, 357.719165]
}
df = pd.DataFrame(data=transaction, index=level_index)
df

It is easy if you only want to select the last row,

df.tail(1)

Turn it into dict

df.tail(1).reset_index().loc[0].to_dict()

### Output
{'code': 'HK.00700',
 'time_key': Timestamp('2022-06-28 10:39:00'),
 'open': 360.8,
 'close': 361.4,
 'high': 361.4,
 'low': 360.4,
 'volume': 116300,
 'upper_band': 361.180835,
 'lower_band': 357.719165}