Question

我有两个pd数据帧：一个包含我使用sklearn.preprocessing StandardScaler缩放的连续变量，另一个包含我使用pd.get_dummies进行虚拟化的分类特征。它们都具有相同数量的行（5443）但具有不同数量的列当我尝试连接这两个数据帧时，按轴= 1（列），Python抛出

ValueError: Shape of passed values is (101, 5936), indices imply (101, 5443)

我检查了数据框的两种形状，我也使用了np.concatenate函数，该函数可以正常工作，但是它会弄乱我对类别特征的伪编码

scaled.shape
Out[378]: (5443, 18)

cat_feats.shape
Out[379]: (5443, 83)

test = pd.concat([scaled,cat_feats],axis=1)
Traceback (most recent call last):

  File "<ipython-input-380-f192aab9181d>", line 1, in <module>
    test = pd.concat([scaled,cat_feats],axis=1)

  File "C:\Users\Marlies\Anaconda3\lib\site-packages\pandas\core\reshape\concat.py", line 213, in concat
    return op.get_result()

  File "C:\Users\Marlies\Anaconda3\lib\site-packages\pandas\core\reshape\concat.py", line 408, in get_result
    copy=self.copy)

  File "C:\Users\Marlies\Anaconda3\lib\site-packages\pandas\core\internals.py", line 5207, in concatenate_block_managers
    return BlockManager(blocks, axes)

  File "C:\Users\Marlies\Anaconda3\lib\site-packages\pandas\core\internals.py", line 3033, in __init__
    self._verify_integrity()

  File "C:\Users\Marlies\Anaconda3\lib\site-packages\pandas\core\internals.py", line 3244, in _verify_integrity
    construction_error(tot_items, block.shape[1:], self.axes)

  File "C:\Users\Marlies\Anaconda3\lib\site-packages\pandas\core\internals.py", line 4608, in construction_error
    passed, implied))

ValueError: Shape of passed values is (101, 5936), indices imply (101, 5443)

结果应为形状为（5443,101）的数据框

当我想根据列连接两个数据帧时，pd.concat引发ValueError

0 个答案: