numpy - 值错误: setting an array element with a sequence with Decision Tree where all the rows have equal elements?

Question

我正在尝试将决策树拟合到特征和标签矩阵中。这是我的代码:

print FEATURES_DATA[0]
print ""
print TARGET[0]
print ""
print np.unique(list(map(len, FEATURES_DATA[0])))

给出以下输出:

[ array([[3, 3, 3, ..., 7, 7, 7],
       [3, 3, 3, ..., 7, 7, 7],
       [3, 3, 3, ..., 7, 7, 7],
       ..., 
       [2, 2, 2, ..., 6, 6, 6],
       [2, 2, 2, ..., 6, 6, 6],
       [2, 2, 2, ..., 6, 6, 6]], dtype=uint8)]

[ array([[31],
       [31],
       [31],
       ..., 
       [22],
       [22],
       [22]], dtype=uint8)]

[463511]

该矩阵实际上包含 463511 个样本。

此后，我运行以下 block :

from sklearn.tree import DecisionTreeClassifier
for i in xrange(5):
    Xtrain=FEATURES_DATA[i]
    Ytrain=TARGET[i]
    clf=DecisionTreeClassifier()
    clf.fit(Xtrain,Ytrain)

这给了我以下错误:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-4-3d8b2a7a3e5f> in <module>()
      4     Ytrain=TARGET[i]
      5     clf=DecisionTreeClassifier()
----> 6     clf.fit(Xtrain,Ytrain)

C:\Users\singhg2\AppData\Local\Enthought\Canopy\User\lib\site-packages\sklearn\tree\tree.pyc in fit(self, X, y, sample_weight, check_input, X_idx_sorted)
    152         random_state = check_random_state(self.random_state)
    153         if check_input:
--> 154             X = check_array(X, dtype=DTYPE, accept_sparse="csc")
    155             if issparse(X):
    156                 X.sort_indices()

C:\Users\singhg2\AppData\Local\Enthought\Canopy\User\lib\site-packages\sklearn\utils\validation.pyc in check_array(array, accept_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, warn_on_dtype, estimator)
    371                                       force_all_finite)
    372     else:
--> 373         array = np.array(array, dtype=dtype, order=order, copy=copy)
    374 
    375         if ensure_2d:

ValueError: setting an array element with a sequence.

我搜索了SO上的其他帖子，发现大多数答案都是矩阵不是完全数字，或者数组在样本之间的长度不同。但是，我的问题不就是这样吗？

有什么帮助吗？

Answer 1

如果打印FEATURES_DATA[0]实际上打印

[ array([[3, 3, 3, ..., 7, 7, 7],
       [3, 3, 3, ..., 7, 7, 7],
       [3, 3, 3, ..., 7, 7, 7],
       ..., 
       [2, 2, 2, ..., 6, 6, 6],
       [2, 2, 2, ..., 6, 6, 6],
       [2, 2, 2, ..., 6, 6, 6]], dtype=uint8)]

那么问题是 FEATURES_DATA[0] 是一个 python 列表，里面有一个 numpy 数组。 (你可以从[和]来理解)

您可以选择列表的第一个(也是唯一的)元素来修复它

from sklearn.tree import DecisionTreeClassifier
for i in xrange(5):
    Xtrain=FEATURES_DATA[i][0]
    Ytrain=TARGET[i][0]
    clf=DecisionTreeClassifier()
    clf.fit(Xtrain,Ytrain)