keras解决多标签分类问题

keras解决多标签分类问题 | DataLearnerAI

from keras.layers import Input,Dense
from keras.models import Sequential

model = Sequential()
model.add(Dense(10, activation="relu", input_shape=(10,)))
model.add(Dense(5))

import numpy as np

def Softmax_sim(x):
    y = np.exp(x)
    return y/np.sum(y)

x = np.array([1.0,2.0,3.0,4.0,1.0])
print(Softmax_sim(x))
#输出：[ 0.03106277  0.08443737  0.22952458  0.6239125   0.03106277]

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

inputs = Input(shape=(10,))
hidden = Dense(units=10,activation='relu')(inputs)
output = Dense(units=5,activation='softmax')(hidden)

z = np.array([-1.0, 5.0, -0.5, 5.0, -0.5 ])
print(Softmax_sim(z))
# 输出为[ 0.00123281  0.49735104  0.00203256  0.49735104  0.00203256]

import numpy as np

def Sigmoid_sim(x):
    return  1 /(1+np.exp(-x))

a = np.array([-1.0, 5.0, -0.5, 5.0, -0.5])
print(Sigmoid_sim(a))
#输出为： [ 0.26894142  0.99330715  0.37754067  0.99330715  0.37754067]

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

inputs = Input(shape=(10,))
hidden = Dense(units=10,activation='relu')(inputs)
output = Dense(units=5,activation='sigmoid')(hidden)
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

keras解决多标签分类问题

DataLearner 官方微信

multi-class classification

mulit-label classification

参考

热门博客