Python Keras - categorical_accuracy 和 sparse_categorical_accuracy 之间的区别
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/44477489/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Keras - Difference between categorical_accuracy and sparse_categorical_accuracy
提问by reindeer
What is the difference between categorical_accuracyand sparse_categorical_accuracyin Keras? There is no hint in the documentation for these metrics, and by asking Dr. Google, I did not find answers for that either.
Kerascategorical_accuracy和sparse_categorical_accuracy在 Keras 中有什么区别？这些指标的文档中没有任何提示，通过询问 Google 博士，我也没有找到答案。
The source code can be found here:
源代码可以在这里找到：
def categorical_accuracy(y_true, y_pred):
return K.cast(K.equal(K.argmax(y_true, axis=-1),
K.argmax(y_pred, axis=-1)),
K.floatx())
def sparse_categorical_accuracy(y_true, y_pred):
return K.cast(K.equal(K.max(y_true, axis=-1),
K.cast(K.argmax(y_pred, axis=-1), K.floatx())),
K.floatx())
采纳答案by Matti Lyra
Looking at the source
看源码
def categorical_accuracy(y_true, y_pred):
return K.cast(K.equal(K.argmax(y_true, axis=-1),
K.argmax(y_pred, axis=-1)),
K.floatx())
def sparse_categorical_accuracy(y_true, y_pred):
return K.cast(K.equal(K.max(y_true, axis=-1),
K.cast(K.argmax(y_pred, axis=-1), K.floatx())),
K.floatx())
categorical_accuracychecks to see if the indexof the maximal true value is equal to the indexof the maximal predicted value.
categorical_accuracy进行检查以查看是否索引的最大真值的等于索引的最大预测值。
sparse_categorical_accuracychecks to see if the maximal true value is equal to the indexof the maximal predicted value.
sparse_categorical_accuracy检查最大真实值是否等于最大预测值的索引。
From Marcin's answer above the categorical_accuracycorresponds to a one-hotencoded vector for y_true.
从上面 Marcin 的回答中，categorical_accuracy对应于 的one-hot编码向量y_true。
回答by Marcin Mo?ejko
So in categorical_accuracyyou need to specify your target (y) as one-hot encoded vector (e.g. in case of 3 classes, when a true class is second class, yshould be (0, 1, 0). In sparse_categorical_accuracyyou need should only provide an integer of the true class (in the case from previous example - it would be 1as classes indexing is 0-based).
因此，categorical_accuracy您需要将目标 ( y)指定为单热编码向量（例如，在 3 个类的情况下，当真正的类是第二类时，y应该是(0, 1, 0)。sparse_categorical_accuracy您只需要提供真实类的整数（在来自上一个示例的案例 -1就像类索引是0基于的一样）。
