Tensorflow%20实战Google深度学习框架 4.2.2 自定义损失函数源代码

时间：2021-09-07 15:40:55
import os

import tab

import tensorflow as tf

from numpy.random import RandomState

print "hello tensorflow 4.1"

batch_size = 8

x = tf.placeholder(tf.float32,shape=(None,2),name='x-input')

y_ = tf.placeholder(tf.float32,shape=(None,1),name='y-input')

w1 = tf.Variable(tf.random_normal([2,1],stddev=1,seed=1))

#w2 = tf.Variable(tf.random_normal([3,1],stddev=1,seed=1))

y = tf.matmul(x,w1)

#a = tf.matmul(x,w1)

#y = tf.matmul(a,w2)

loss_less = 10

loss_more = 1

loss = tf.reduce_sum(tf.where(tf.greater(y,y_),(y-y_)*loss_more,(y_-y)*loss_less))

train_step = tf.train.AdamOptimizer(0.001).minimize(loss)

rdm = RandomState(1)

dataset_size = 128

X = rdm.rand(dataset_size,2)

Y = [[x1 + x2 +rdm.rand()/10.0-0.05] for (x1 ,x2 ) in X]

with tf.Session() as sess:

    init_op = tf.global_variables_initializer()

    sess.run(init_op)

    print sess.run(w1)

    STEPS = 5000

    for i in range(STEPS):

        start = (i * batch_size) % dataset_size

        end = min(start+batch_size,dataset_size)

        sess.run(train_step, feed_dict = {x: X[start:end], y_: Y[start:end]} )

        print sess.run(w1)

print "end "

　　

相关文章