tensorflow2.0——交叉熵损失应用

import tensorflow as tf

a = tf.losses.categorical_crossentropy([0,1,0,0],[0.25,0.25,0.25,0.25],from_logits=True)     #   前一个参数为标记值，后一个参数为预测值,最后一个参数设为True，输出就不用做softmax
print('a:',a)
b = tf.losses.categorical_crossentropy([0,1,0,0],[0.1,0.1,0.1,0.7])     #   前一个参数为标记值，后一个参数为预测值
print('b:',b)
c = tf.losses.categorical_crossentropy([0,1,0,0],[0.1,0.7,0.1,0.1])     #   前一个参数为标记值，后一个参数为预测值
print('c:',c)
d = tf.losses.categorical_crossentropy([0,1,0,0],[0,0.7,0,0.3])     #   前一个参数为标记值，后一个参数为预测值
print('d:',d)
e = tf.losses.categorical_crossentropy([0,1,0,0],[0.02,0.9,0.03,0.05])     #   前一个参数为标记值，后一个参数为预测值
print('e:',e)
f = tf.losses.categorical_crossentropy([1,0],[0.9,0.1])     #   前一个参数为标记值，后一个参数为预测值
print('f:',f)