tensorflow实现逻辑斯特回归（logistic regression）

tensorflow实现逻辑斯特回归（logistic regression）

import tensorflow as tf
import numpy as np

from tensorflow.examples.tutorials import mnist

mnist = mnist.input_data.read_data_sets('./data/mnist',one_hot=True)

x = tf.placeholder(dtype=tf.float32,shape=[None,784],name='input')
y = tf.placeholder(dtype=tf.float32,shape=[None,10],name='output')

w = tf.Variable(tf.random_normal([784,10],0,0.05,dtype=tf.float32),name='weights')
b = tf.Variable(tf.random_normal([1,10],dtype=tf.float32),name='bias')

logits = tf.add(tf.matmul(x,w),b)

accuracy = tf.reduce_mean(tf.cast(tf.equal(tf.argmax(logits,1),tf.argmax(y,1)),tf.float32))

lr = 0.01
epochs = 100
batch_size = 64
loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits,labels=y),name='loss')
optimizer = tf.train.AdamOptimizer(learning_rate=lr).minimize(loss)

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    writer = tf.summary.FileWriter('./graphs/logistic_reg',sess.graph)
    n_batch = mnist.train.num_examples//batch_size
    for epoch in range(epochs):
        train_loss = 0
        train_acc = 0
        for _ in range(n_batch):
            xs,ys = mnist.train.next_batch(batch_size)
            _,batch_loss,batch_acc = sess.run([optimizer,loss,accuracy],feed_dict={x:xs,y:ys})
            train_loss += batch_loss
            train_acc += batch_acc

        if epoch % 10 == 0:
            val_loss,val_acc = sess.run([loss,accuracy],feed_dict={x:mnist.validation.images,y:mnist.validation.labels})

            print('epoch {} train loss = {},train acc {},the val loss={},val acc {}'\
                .format(epoch,train_loss/n_batch,train_acc/n_batch,val_loss,val_acc))

    test_loss,test_acc= sess.run([loss,accuracy],feed_dict={x:mnist.test.images,y:mnist.test.labels})

    print('the model on test loss is {},the accuracy is {}'.format(test_loss,test_acc))
    writer.close()