sklearn管道转换ValueError期望值不等于训练值 [英] sklearn pipeline transform ValueError that Expected Value is not equal to Trained Value

问题描述

请提供以下函数，让我遇到ValueError错误：使用剩余关键字时列排序必须相等，以适合拟合和变换

Can you please help me to with the following function where I got the error of ValueError: Column ordering must be equal for fit and for transform when using the remainder keyword

（该函数在我保存在GCP存储中的腌制sklearn管道上调用。）

(The function is called on a pickled sklearn pipeline that I had saved in GCP Storage.)

错误：

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-192-c6a8bc0ab221> in <module>
----> 1 safety_project_lite(request)

<ipython-input-190-24c565131f14> in safety_project_lite(request)
     31 
     32     df_resp = pd.DataFrame(data=request_data)
---> 33     response = loaded_model.predict(df_resp)
     34 
     35     output = {"Safety Rating": response[0]}

~/.local/lib/python3.5/site-packages/sklearn/utils/metaestimators.py in <lambda>(*args, **kwargs)
    114 
    115         # lambda, but not partial, allows help() to work with update_wrapper
--> 116         out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)
    117         # update the docstring of the returned function
    118         update_wrapper(out, self.fn)

~/.local/lib/python3.5/site-packages/sklearn/pipeline.py in predict(self, X, **predict_params)
    417         Xt = X
    418         for _, name, transform in self._iter(with_final=False):
--> 419             Xt = transform.transform(Xt)
    420         return self.steps[-1][-1].predict(Xt, **predict_params)
    421 

~/.local/lib/python3.5/site-packages/sklearn/compose/_column_transformer.py in transform(self, X)
    581             if (n_cols_transform >= n_cols_fit and
    582                     any(X.columns[:n_cols_fit] != self._df_columns)):
--> 583                 raise ValueError('Column ordering must be equal for fit '
    584                                  'and for transform when using the '
    585                                  'remainder keyword')

ValueError: Column ordering must be equal for fit and for transform when using the remainder keyword

Code：

def safety_project_lite_beta(request):
   client = storage.Client(request.GCP_Project)
   bucket = client.get_bucket(request.GCP_Bucket)
   blob = bucket.blob(request.GCP_Path)
   model_file = BytesIO()
   blob.download_to_file(model_file)
   loaded_model = pickle.loads(model_file.getvalue())

   request_data = {'A': [request.A],
   'B': [request.B],
   'C': [request.C],
   'D': [request.D],
   'E': [request.E],
   'F': [request.F]}

   df_resp = pd.DataFrame(data=request_data)
   response = loaded_model.predict(df_resp)

   output = {"Rating": response[0]}

   return output

推荐答案

该模型只能预测您提供的数据的结构是否与经过训练的结构相同。

The model can only predict if the data you feed it is of the same structure as it has been trained on.

要强制以下事实，即 df_resp 与 X_train 具有相同的列，在建立数据框时传递其列的列表：

To force the fact that df_resp has the same columns as X_train, pass a list of its columns along when building the dataframe:

df_resp = pd.DataFrame(request_data, columns=X_train.columns)

如果该变量由于某种原因不可用，则可以对其进行腌制列列表（ X_train.columns ）并在以后使用：

If that variable is for some reason not available, you could pickle its column list (X_train.columns) and use it later:

loaded_cols = pickle.loads([...])
df_resp = pd.DataFrame(data=request_data, columns=loaded_cols)

这样可以确保工作流更加动态，例如，您可以更轻松地添加列

This ensures a more dynamic workflow where you could add columns more easily for example.

这篇关于sklearn管道转换ValueError期望值不等于训练值的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！