Python中某些函数的贬值：数据帧的真值不明确

Question

当我在这部分遇到这个错误时，我正在训练各种ANN。不知道有没有什么功能被贬值了

def analyze_results(rainfall_data, test_rainfall_data, name, flag=False):
    optimized_params = rainfall_data.loc[rainfall_data.RMSE.argmin]
    future_steps = optimized_params.future_steps
    forecast_values = optimized_params[-1*int(future_steps):]
    y_true = test_rainfall_data.iloc[:int(future_steps)]
    forecast_values.index = y_true.index
    
    print('=== Best parameters of ' + name + ' ===\n')
    if (name == 'FNN' or name == 'LSTM'):
        model = create_NN(optimized_params.look_back, 
                          optimized_params.hidden_nodes, 
                          optimized_params.output_nodes)
        print('Input nodes(p): ' + str(optimized_params.look_back))
        print('Hidden nodes: ' + str(optimized_params.hidden_nodes))
        print('Output nodes: ' + str(optimized_params.output_nodes))
    elif (name == 'TLNN'):
        model = create_NN(len(optimized_params.look_back_lags), 
                          optimized_params.hidden_nodes, 
                          optimized_params.output_nodes)
        s = ''
        for i in optimized_params.look_back_lags:
            s = s+' '+str(i)
        print('Look back lags: ' + s)
        print('Hidden nodes: ' + str(optimized_params.hidden_nodes))
        print('Output nodes: ' + str(optimized_params.output_nodes))
    elif (name == 'SANN'):
        model = create_NN(optimized_params.seasonal_period, 
                          optimized_params.hidden_nodes, 
                          optimized_params.seasonal_period)
        print('Input nodes(s): ' + str(optimized_params.seasonal_period))
        print('Hidden nodes: ' + str(optimized_params.hidden_nodes))
        print('Output nodes: ' + str(optimized_params.seasonal_period))
        
    print('Number of epochs: ' + str(optimized_params.epochs))
    print('Batch size: ' + str(optimized_params.batch_size))
    print('Number of future steps forecasted: ' + str(optimized_params.future_steps))
    print('Mean Squared Error(MSE): ' + str(optimized_params.MSE))
    print('Mean Absolute Error(MAE): ' + str(optimized_params.MAE))
    print('Root Mean Squared Error(RMSE): ' + str(optimized_params.RMSE))
    print('\n\n')

错误显示在这一行

optimized_params = rainfall_data.loc[(rainfall_data.RMSE.argmin)]

错误显示为

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_12964\1785142711.py in ?()
      9 
     10 # look_back, hidden_nodes, output_nodes, epochs, batch_size, future_steps
     11 parameters_LSTM = [[1,2,3,4,5,6,7,8,9,10,11,12,13], [3,4,5,6], [1], [300], [20], [future_steps]]
     12 
---> 13 RMSE_info = compare_ANN_methods(rainfall_data, test_rainfall_data, scaler, parameters_FNN, parameters_TLNN, parameters_SANN, parameters_LSTM, future_steps)

~\AppData\Local\Temp\ipykernel_12964\2478982653.py in ?(rainfall_data, test_rainfall_data, scaler, parameters_FNN, parameters_TLNN, parameters_SANN, parameters_LSTM, future_steps)
      1 def compare_ANN_methods(rainfall_data, test_rainfall_data, scaler, parameters_FNN, parameters_TLNN, parameters_SANN, parameters_LSTM, future_steps):
      2 
      3     information_FNN_df = get_accuracies_FNN(rainfall_data, test_rainfall_data, parameters_FNN, scaler)
----> 4     optimized_params_FNN = analyze_results(information_FNN_df, test_rainfall_data, 'FNN')
      5 
      6     information_TLNN_df = get_accuracies_TLNN(rainfall_data, test_rainfall_data, parameters_TLNN, scaler)
      7     optimized_params_TLNN = analyze_results(information_TLNN_df, test_rainfall_data, 'TLNN')

~\AppData\Local\Temp\ipykernel_12964\4019196368.py in ?(rainfall_data, test_rainfall_data, name, flag)
      1 def analyze_results(rainfall_data, test_rainfall_data, name, flag=False):
----> 2     optimized_params = rainfall_data.loc[(rainfall_data.RMSE.argmin)]
      3     future_steps = optimized_params.future_steps
      4     forecast_values = optimized_params[-1*int(future_steps):]
      5     y_true = test_rainfall_data.iloc[:int(future_steps)]

~\anaconda3\Lib\site-packages\pandas\core\indexing.py in ?(self, key)
   1185         else:
   1186             # we by definition only have the 0th axis
   1187             axis = self.axis or 0
   1188 
-> 1189             maybe_callable = com.apply_if_callable(key, self.obj)
   1190             maybe_callable = self._check_deprecated_callable_usage(key, maybe_callable)
   1191             return self._getitem_axis(maybe_callable, axis=axis)

~\anaconda3\Lib\site-packages\pandas\core\common.py in ?(maybe_callable, obj, **kwargs)
    380     obj : NDFrame
    381     **kwargs
    382     """
    383     if callable(maybe_callable):
--> 384         return maybe_callable(obj, **kwargs)
    385 
    386     return maybe_callable

~\anaconda3\Lib\site-packages\pandas\core\base.py in ?(self, axis, skipna, *args, **kwargs)
    765     def argmin(
    766         self, axis: AxisInt | None = None, skipna: bool = True, *args, **kwargs
    767     ) -> int:
    768         delegate = self._values
--> 769         nv.validate_minmax_axis(axis)
    770         skipna = nv.validate_argmin_with_skipna(skipna, args, kwargs)
    771 
    772         if isinstance(delegate, ExtensionArray):

~\anaconda3\Lib\site-packages\pandas\compat\numpy\function.py in ?(axis, ndim)
    395     ValueError
    396     """
    397     if axis is None:
    398         return
--> 399     if axis >= ndim or (axis < 0 and ndim + axis < 0):
    400         raise ValueError(f"`axis` must be fewer than the number of dimensions ({ndim})")

~\anaconda3\Lib\site-packages\pandas\core\generic.py in ?(self)
   1575     @final
   1576     def __nonzero__(self) -> NoReturn:
-> 1577         raise ValueError(
   1578             f"The truth value of a {type(self).__name__} is ambiguous. "
   1579             "Use a.empty, a.bool(), a.item(), a.any() or a.all()."
   1580         )

ValueError: The truth value of a DataFrame is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

Answer 1

您遇到的错误是由于您尝试使用

RMSE

数据框中的

rainfall_data

访问提供的行，使用

argmin

检索具有最小值

RMSE 

的行的正确方法，如下所示：

optimized_params = rainfall_data.loc[rainfall_data['RMSE'}.idxmin()]

这就是为什么：

问题是

argmin

是一种较旧的方法，它返回最小值的索引。 而

idxmin

是一个现代的 pandas，它直接返回具有最小值的索引标签。 https://pandas.pydata.org/docs/reference/api/pandas.Series.argmin.html

我能给出的关于错误消息的唯一解释

ValueError: The truth value of a DataFrame is ambiguous

特别表明pandas无法确定如何在布尔上下文中解释整个DataFrame

(rainfall_data)

。这种混乱源于尝试使用

.argmin

，这不是查找 pandas 系列

(like rainfall_data['RMSE'])

中最小值索引的有效方法。