分类器：轴 1 超出维度 1 数组的范围

Question

我正在使用 ScikitlearnRandomForestClassifier 如下。

from art.estimators.classification.scikitlearn import ScikitlearnRandomForestClassifier as SRFC
from sklearn.ensemble import RandomForestClassifier as RFC

model_rand_forest = SRFC(RFC(n_estimators=500,
                             max_depth=45,
                             criterion='entropy',
                             random_state=32))
model_rand_forest = model_rand_forest.fit(x_train, y_train)

我遇到了以下错误。 如果需要， y_train是一系列大小(70540,) ， x_train是大小(70540, 128) 。

error:
 File "<ipython-input-18-c410427d7973>", line 1, in <module>
    model_rand_forest = model_rand_forest.fit(x_train , y_train)

  File "E:\Anaconda3\lib\site-packages\art\estimators\classification\classifier.py", line 71, in replacement_function
    return fdict[func_name](self, *args, **kwargs)

  File "E:\Anaconda3\lib\site-packages\art\estimators\classification\scikitlearn.py", line 138, in fit
    y_preprocessed = np.argmax(y_preprocessed, axis=1)

  File "<__array_function__ internals>", line 6, in argmax

  File "E:\Anaconda3\lib\site-packages\numpy\core\fromnumeric.py", line 1186, in argmax
    return _wrapfunc(a, 'argmax', axis=axis, out=out)

  File "E:\Anaconda3\lib\site-packages\numpy\core\fromnumeric.py", line 61, in _wrapfunc
    return bound(*args, **kwds)

AxisError: axis 1 is out of bounds for array of dimension 1

Answer 1

您必须向 scikit-learn 模型的 ART 估计器提供使用形状 (nb_samples, nb_classes) 编码的标签 y。 您可以使用此链接将您的标签置于 one-hot 格式中。 https://machinelearningmastery.com/how-to-one-hot-encode-sequence-data-in-python/