Tensorflow简单复现softmax模型对于mnist数据集分类

最新推荐文章于 2023-09-13 12:19:26 发布

原创最新推荐文章于 2023-09-13 12:19:26 发布 · 527 阅读

0 ·

CC 4.0 BY-SA版权

DL 专栏收录该内容

6 篇文章

订阅专栏

本文通过使用TensorFlow实现了一个简单的MNIST手写数字识别模型，并对比了GradientDescentOptimizer和AdamOptimizer两种优化器的效果。在输入数据定义、模型变量定义、模型定义及成本函数定义的基础上，展示了如何用GradientDescentOptimizer进行训练。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

代码中对于GradientDescentOptimizer和AdamOptimizer进行了简单比较。

import tensorflow as tf 
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

# Input data definition
x = tf.placeholder(tf.float32,[None,784])
y = tf.placeholder(tf.float32,[None,10])
# Model variable definition
W = tf.Variable(tf.zeros([784,10]))
b = tf.Variable(tf.zeros([10]))
# Define model
y_predict = tf.matmul(x,W)+b
# Define cost function 
cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels = y,logits = y_predict))
# Thinking1 - What will happen here,if i use adam optimizer?
#optimizer = tf.train.AdamOptimizer(1e-4)
optimizer = tf.train.GradientDescentOptimizer(0.5)
train_step = optimizer.minimize(cross_entropy)

sess = tf.InteractiveSession()
global_initial = tf.global_variables_initializer()
sess.run(global_initial)

for i in range(1000):
	batch = mnist.train.next_batch(100)
	sess.run(train_step,feed_dict={x:batch[0],y:batch[1]})

correct_prediction = tf.equal(tf.argmax(y_predict,1),tf.argmax(y,1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
print(sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels}))

print(sess.run(b))