Python Keras 如何处理多标签分类？
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/44164749/
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
How does Keras handle multilabel classification?
提问by user798719
I am unsure how to interpret the default behavior of Keras in the following situation:
我不确定在以下情况下如何解释 Keras 的默认行为：
My Y (ground truth) was set up using scikit-learn's MultilabelBinarizer().
我的 Y（基本事实）是使用 scikit-learn 的MultilabelBinarizer() 设置的。
Therefore, to give a random example, one row of my ycolumn is one-hot encoded as such:
[0,0,0,1,0,1,0,0,0,0,1].
因此，为了得到无规例如，我的一排y列是独热编码为这样的：
[0,0,0,1,0,1,0,0,0,0,1]。
So I have 11 classes that could be predicted, and more than one can be true; hence the multilabel nature of the problem. There are three labels for this particular sample.
所以我有 11 个可以预测的类，并且不止一个类是真实的；因此问题的多标签性质。这个特定样本有三个标签。
I train the model as I would for a non multilabel problem (business as usual) and I get no errors.
我像处理非多标签问题一样训练模型（照常营业），并且没有出现任何错误。
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.optimizers import SGD
model = Sequential()
model.add(Dense(5000, activation='relu', input_dim=X_train.shape[1]))
model.add(Dropout(0.1))
model.add(Dense(600, activation='relu'))
model.add(Dropout(0.1))
model.add(Dense(y_train.shape[1], activation='softmax'))
sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy',
optimizer=sgd,
metrics=['accuracy',])
model.fit(X_train, y_train,epochs=5,batch_size=2000)
score = model.evaluate(X_test, y_test, batch_size=2000)
score
What does Keras do when it encounters my y_trainand sees that it is "multi" one-hot encoded, meaning there is more than one 'one' present in each row of y_train? Basically, does Keras automatically perform multilabel classification? Any differences in the interpretation of the scoring metrics?
当 Keras 遇到 myy_train并看到它是“多”单热编码时，它会做什么，这意味着每一行中存在多个“一个” y_train？基本上，Keras 会自动执行多标签分类吗？对评分指标的解释有什么不同吗？
回答by frankyjuang
In short
简而言之
Don't use softmax.
不要使用softmax.
Use sigmoidfor activation of your output layer.
使用sigmoid你的输出层的激活。
Use binary_crossentropyfor loss function.
使用binary_crossentropy的损失函数。
Use predictfor evaluation.
使用predict进行评估。
Why
为什么
In softmaxwhen increasing score for one label, all others are lowered (it's a probability distribution). You don't want that when you have multiple labels.
在softmax增加一个标签的分数时，所有其他标签都会降低（这是一种概率分布）。当您有多个标签时，您不希望这样。
Complete Code
完整代码
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.optimizers import SGD
model = Sequential()
model.add(Dense(5000, activation='relu', input_dim=X_train.shape[1]))
model.add(Dropout(0.1))
model.add(Dense(600, activation='relu'))
model.add(Dropout(0.1))
model.add(Dense(y_train.shape[1], activation='sigmoid'))
sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='binary_crossentropy',
optimizer=sgd)
model.fit(X_train, y_train, epochs=5, batch_size=2000)
preds = model.predict(X_test)
preds[preds>=0.5] = 1
preds[preds<0.5] = 0
# score = compare preds and y_test
