# _*_ encoding=utf8 _*_
import tensorflow as tf
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("data/", one_hot=True)
# 设置学习率,
learning_rate = 0.01
training_epochs = 25
batch_size = 100
display_step = 1
#图片都是28*28 = 784的 预测为0-9 10个数字 10分类
x = tf.placeholder(tf.float32, [None, 784])
y = tf.placeholder(tf.float32, [None, 10])
# 初始化参数的值
W = tf.Variable(tf.zeros([28*28, 10]),name="W")
b = tf.Variable(tf.zeros([10]))
# 前向传播
pred = tf.nn.softmax(tf.matmul(x, W) + b)
# 使用交叉熵作为损失函数
cost = tf.reduce_mean(-tf.reduce_sum(y*tf.log(pred), reduction_indices=1))
# 梯度下降求最佳的cost
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)
init = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init)
print(W) #<tf.Variable 'W:0' shape=(784, 10) dtype=float32_ref>
for epoch in range(training_epochs):
# 得到数据总的batch
total_batch = int(mnist.train.num_examples / batch_size)
# print(total_batch) #550 个batch 55000张训练数据
# 每一训练 都去求得平均的cost
avg_cost = 0
#batch训练 遍历所以得batch
for i in range(total_batch):
batch_x,batch_y = mnist.train.next_batch(batch_size)
# 训练参数
_, c = sess.run([optimizer, cost], feed_dict={x: batch_x,y: batch_y})
avg_cost += c / total_batch
# 计算每一次迭代的cost
if (epoch + 1) % display_step == 0:
print("Epoch:", '%04d' % (epoch + 1), "cost=", "{:.9f}".format(avg_cost))
#评估模型 训练出来的结果tf.argmax(pred, 1) 找10个数最大的那个 和真实值y对比
correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
# 计算准确率
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
# 使用mnist的test数据进行评估
# 首先执行之前的所有必要的操作来产生这个计算这个tensor需要的输入,然后通过这些输入产生这个tensor。
print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
Tensorflow撸代码之2逻辑回归
最新推荐文章于 2023-12-18 17:30:00 发布