在Keras中,如何在权重矩阵的每一行上应用softmax函数?

时间:2018-11-13 03:41:45

标签: python machine-learning keras keras-layer softmax

from keras.models import Model
from keras.models import Input
from keras.layers import Dense

a = Input(shape=(3,))
b = Dense(2, use_bias=False)(a)
model = Model(inputs=a, outputs=b)

假设以上代码中Dense层的权重为[[2, 3], [3, 1], [-1, 1]]。如果我们将[[2, 1, 3]]作为model的输入,则输出将为:

no softmax

但是我想将softmax函数应用于Dense层的每一行,以便输出为:

with softmax

我该怎么做?

1 个答案:

答案 0 :(得分:1)

一种实现所需内容的方法是通过对Dense层进行子类化并覆盖其call方法来定义自定义层:

from keras import backend as K

class CustomDense(Dense):
    def __init__(self, units, **kwargs):
        super(CustomDense, self).__init__(units, **kwargs)

    def call(self, inputs):
        output = K.dot(inputs, K.softmax(self.kernel, axis=-1))
        if self.use_bias:
            output = K.bias_add(output, self.bias, data_format='channels_last')
        if self.activation is not None:
            output = self.activation(output)
        return output

进行测试以确保其正常工作

model = Sequential()
model.add(CustomDense(2, use_bias=False, input_shape=(3,)))

model.compile(loss='mse', optimizer='adam')

import numpy as np

w = np.array([[2,3], [3,1], [1,-1]])
inp = np.array([[2,1,3]])

model.layers[0].set_weights([w])
print(model.predict(inp))

# output
[[4.0610714 1.9389288]]

使用numpy进行验证:

soft_w = np.exp(w) / np.sum(np.exp(w), axis=-1, keepdims=True)
print(np.dot(inp, soft_w))

[[4.06107115 1.93892885]]