Sklearn Lasso预测ndarry输出问题

时间:2015-08-09 01:18:18

标签: machine-learning scikit-learn classification

我试图使用sklearn输出预测输出,但实际的ndarray看起来与Lasso的输出不同,因为我无法计算错误。我有什么事吗?我希望Y_PRED与Y_ACT看起来相同。

Y_PRED:
[[-0.2345764   2.95901652 -0.49351205 ..., -0.4138885  -0.51545272
  -0.43058412]]
Y_ACT :
[[-0.72284503]
 [ 3.31750686]
 [-0.42046417]
 ...,
 [-0.75198836]
 [-0.80438594]]

以下是示例代码:

  x_df = TR_SET_FEATURE
  y_df = TR_SET_TARGET
  for train_index, test_index in kf:
     x_train = x_df.iloc[train_index, :].values
     x_test = x_df.iloc[test_index, :].values

     y_train = y_df.iloc[train_index].values
     y_test = y_df.iloc[test_index].values

     regr = ElasticNet(alpha=a)
     regr.fit(x_train, y_train)
     y_pred = regr.predict(x_test)

     print "y_pred: " ,y_pred.shape()
     print "y_act: " ,y_test.shape()


     pred_error = y_pred - y_test

输出:
        y_pred:       文件" dataframe.py",第700行,in         打印" y_pred:" ,y_pred.shape()

0 个答案:

没有答案