Question

我尝试计算不同深度的决策树的测试和训练错误。

train_error = []
test_error = []    
for i in range (3,21):
    X_train, X_test, y_train, y_test = train_test_split(womendata, womeny, test_size=0.4, random_state=1 )
    decitiontree = tree.DecisionTreeClassifier(criterion='gini', splitter='best', max_depth=i, class_weight = 'balanced', min_samples_split=i)
    clf = decitiontree.fit(X_train, y_train)
    train_error.append( 1 -  clf.score(X_train, y_train)  )     
    test_error.append( 1 -  clf.score(X_test, y_test)  )

在python 3中我收到错误：

Traceback (most recent call last):
  File "<stdin>", line 4, in <module>
  File "/usr/local/lib/python3.4/dist-packages/sklearn/tree/tree.py", line 154, in fit
    X = check_array(X, dtype=DTYPE, accept_sparse="csc")
  File "/usr/local/lib/python3.4/dist-packages/sklearn/utils/validation.py", line 398, in check_array
    _assert_all_finite(array)
  File "/usr/local/lib/python3.4/dist-packages/sklearn/utils/validation.py", line 54, in _assert_all_finite
    " or a value too large for %r." % X.dtype)

ValueError: Input contains NaN, infinity or a value too large for dtype('float32').

womendata en women y的长度相同，并且集合中没有丢失的数据。

Answer 1

从您提供包含无效值的数据数组的错误。

ValueError：输入包含 NaN，无穷大或值太大 D型（ 'FLOAT32'）。

请检查您的数据是否有效意义：

在womendata或womeny上没有NaN值
没有关于womendata或womeny的Inf值
值在float32 min和float32 max

您可以使用以下代码：

import numpy as np
info = np.finfo(np.float64)

for x in [womendata, womeny]:
    assert np.all(x <= info.max) and np.all(x >= info.min), 'not all values in range'
    assert np.all(x != np.inf) and np.all(x != -np.inf), 'data contains infinity value'
    assert np.all(x is not np.nan), 'data contains Nan value'

不同长度的决策树错误

1 个答案: