直接上代码:(原理或细节不懂百度或查书)
inference.py
import tensorflow as tf
INPUT_NODE = 784
OUTPUT_NODE = 10
IMAGE_SIZE = 28
NUM_CHANNELS = 1
NUM_LABELS = 10
CONV1_DEEP = 32
CONV1_SIZE = 5
CONV2_DEEP = 64
CONV2_SIZE = 5
FC_SIZE = 512
def inference(input_tensor, train, regularizer):
with tf.variable_scope('layer1-conv1'):
conv1_weights = tf.get_variable(
"weight", [CONV1_SIZE, CONV1_SIZE, NUM_CHANNELS, CONV1_DEEP],
initializer=tf.truncated_normal_initializer(stddev=0.1))
conv1_biases = tf.get_variable("bias", [CONV1_DEEP], initializer=tf.constant_initializer(0.0))
conv1 = tf.nn.conv2d(input_tensor, conv1_weights, strides=[1, 1, 1, 1], padding='SAME')
relu1 = tf.nn.relu(tf.nn.bias_add(conv1, conv1_biases))
with tf.name_scope("layer2-pool1"):
pool1 = tf.nn.max_pool(relu1, ksize = [1,2,2,1],strides=[1,2,2,1],padding="SAME")
with tf.variable_scope("layer3-conv2"):
conv2_weights = tf.get_variable(
"weight", [CONV2_SIZE, CONV2_SIZE, CONV1_DEEP, CONV2_DEEP],
initializer=tf.truncated_normal_initializer(stddev=0.1))
conv2_biases = tf.get_variable("bias", [CONV2_DEEP], initializer=tf.constant_initializer(0.0))
conv2 = tf.nn.conv2d(pool1, conv2_weights, strides=[1, 1, 1, 1], padding='SAME')
relu2 = tf.nn.relu(tf.nn.bias_add(conv2, conv2_biases))
with tf.name_scope("layer4-pool2"):
pool2 = tf.nn.max_pool(relu2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
pool_shape = pool2.get_shape().as_list()
nodes = pool_shape[1] * pool_shape[2] * pool_shape[3]
reshaped = tf.reshape(pool2, [pool_shape[0], nodes])
with tf.variable_scope('layer5-fc1'):
fc1_weights = tf.get_variable("weight", [nodes, FC_SIZE],
initializer=tf.truncated_normal_initializer(stddev=0.1))
if regularizer != None: tf.add_to_collection('losses', regularizer(fc1_weights))
fc1_biases = tf.get_variable("bias", [FC_SIZE], initializer=tf.constant_initializer(0.1))
fc1 = tf.nn.relu(tf.matmul(reshaped, fc1_weights) + fc1_biases)
if train: fc1 = tf.nn.dropout(fc1, 0.5)
with tf.variable_scope('layer6-fc2'):
fc2_weights = tf.get_variable("weight", [FC_SIZE, NUM_LABELS],
initializer=tf.truncated_normal_initializer(stddev=0.1))
if regularizer != None: tf.add_to_collection('losses', regularizer(fc2_weights))
fc2_biases = tf.get_variable("bias", [NUM_LABELS], initializer=tf.constant_initializer(0.1))
logit = tf.matmul(fc1, fc2_weights) + fc2_biases
return logit
train.py
import tensorflow as tf
import os
from tensorflow.examples.tutorials.mnist import input_data
import numpy as np
# 加载函数
import inference
# 配置神经网络参数
BATCH_SIZE = 100
LEARNING_RATE_BASE = 0.1 #初始学习率的设置将会严重影响神经网络的收敛速度!!!
LEARNING_RATE_DECAY = 0.99
REGULARIZATION_RATE = 0.0001
TRAINING_STEPS = 10000
MOVING_AVERAGE_DECAY = 0.99
# 模型保存路径和文件名
MODEL_SAVE_PATH = "/path/to/model"
MODEL_NAME = "model.ckpt"
def train(mnist):
# 定义输入输出的placeholder
# x = tf.placeholder(tf.float32, [None, mnist_inference.INPUT_NODE], name='x-input')
x = tf.placeholder(tf.float32, [BATCH_SIZE,
inference.IMAGE_SIZE,
inference.IMAGE_SIZE,
inference.NUM_CHANNELS],
name='x-input')
y_ = tf.placeholder(tf.float32, [BATCH_SIZE, inference.OUTPUT_NODE], name='y-input')
# 定义正则化
regularizer = tf.contrib.layers.l2_regularizer(REGULARIZATION_RATE)
# 使用前向传播
y = inference.inference(x, True, regularizer)
global_step = tf.Variable(0, trainable=False)
# 滑动平均
variable_averages = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
variable_averages_op = variable_averages.apply(tf.trainable_variables())
# print(tf.trainable_variables())
# [<tf.Variable 'layer1-conv1/weight:0' shape=(5, 5, 1, 32) dtype=float32_ref>,
# <tf.Variable 'layer1-conv1/bias:0' shape=(32,) dtype=float32_ref>,
# <tf.Variable 'layer3-conv2/weight:0' shape=(5, 5, 32, 64) dtype=float32_ref>,
# <tf.Variable 'layer3-conv2/bias:0' shape=(64,) dtype=float32_ref>,
# <tf.Variable 'layer5-fc1/weight:0' shape=(3136, 512) dtype=float32_ref>,
# <tf.Variable 'layer5-fc1/bias:0' shape=(512,) dtype=float32_ref>,
# <tf.Variable 'layer6-fc2/weight:0' shape=(512, 10) dtype=float32_ref>,
# <tf.Variable 'layer6-fc2/bias:0' shape=(10,) dtype=float32_ref>]
# 损失函数
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=tf.argmax(y_, 1), logits=y)
cross_entropy_mean = tf.reduce_mean(cross_entropy)
loss = cross_entropy_mean + tf.add_n(tf.get_collection('losses'))
# print(tf.get_collection('losses'))
# #[<tf.Tensor 'layer5-fc1/l2_regularizer:0' shape=() dtype=float32>,
# <tf.Tensor 'layer6-fc2/l2_regularizer:0' shape=() dtype=float32>]
# 学习率
learning_rate = tf.train.exponential_decay(LEARNING_RATE_BASE,
global_step,
mnist.train.num_examples / BATCH_SIZE,
LEARNING_RATE_DECAY)
# 优化算法
train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_step)
with tf.control_dependencies([train_step, variable_averages_op]):
train_op = tf.no_op(name="train")
# 持久化
saver = tf.train.Saver()
# config = tf.ConfigProto() config=config
# config.gpu_options.allow_growth = True
with tf.Session() as sess:
tf.global_variables_initializer().run()
for i in range(TRAINING_STEPS):
xs, ys = mnist.train.next_batch(BATCH_SIZE)
# 调整为四维矩阵
reshaped_xs = np.reshape(xs, [BATCH_SIZE,
inference.IMAGE_SIZE,
inference.IMAGE_SIZE,
inference.NUM_CHANNELS])
# 运行
_, loss_valuue, step = sess.run([train_op, loss, global_step], feed_dict={x: reshaped_xs, y_: ys})
# 每1000轮保存一次模型
if i % 1000 == 0:
print("After %d training step(s), loss on training batch is %g." % (step, loss_valuue))
saver.save(sess, os.path.join(MODEL_SAVE_PATH, MODEL_NAME), global_step=global_step)
def main(argv=None):
mnist = input_data.read_data_sets("C:/path/to/MNIST_data", one_hot=True)
train(mnist)
if __name__ == '__main__':
tf.app.run()
最后是测试模型在验证上表现的代码
evaluate.py
# -*- coding: utf-8 -*-
import time
import numpy as np
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
# 加载mnist_inference.py和mnist_train.py中定义的常量和函数。
import inference
import train
# 每10秒加载一次最新的模型,并在测试数据上测试最新模型的正确率。
EVAL_INTERVAL_SECS = 30
def evaluate(mnist):
with tf.Graph().as_default() as g:
# 定义输入输出的格式。
x = tf.placeholder(tf.float32,[
mnist.validation.num_examples,
train.inference.IMAGE_SIZE,
train.inference.IMAGE_SIZE,
train.inference.NUM_CHANNELS],
name='x-input')
reshaped_xs = np.reshape(mnist.validation.images,(mnist.validation.num_examples,
inference.IMAGE_SIZE,
inference.IMAGE_SIZE,
inference.NUM_CHANNELS))
y_ = tf.placeholder(
tf.float32, [None, inference.OUTPUT_NODE], name='y-input')
validate_feed = {x:reshaped_xs,
y_:mnist.validation.labels}
# 直接通过调用封装好的函数来计算前向传播的结果。因为测试时不关注正则化损失的值,
# 所以这里用于计算正则化损失的函数被设置为None。
y = inference.inference(x,False,None)
# 使用前向传播的结果计算正确率。如果需要对未知的样例进行分类,那么使用
# tf.argmax(y, 1)就可以得到输入样例的预测类别了。
correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
# 通过变量重命名的方式来加载模型,这样在前向传播的过程中就不需要调用求滑动平均
# 的函数来获取平均值了。这使得我们可以完全共用mnist_inference.py中定义的
# 前向传播过程。
variable_averages = tf.train.ExponentialMovingAverage(
train.MOVING_AVERAGE_DECAY)
variables_to_restore = variable_averages.variables_to_restore()
saver = tf.train.Saver(variables_to_restore)
# 每隔EVAL_INTERVAL_SECS秒调用一次计算正确率的过程以检测训练过程中正确率的# 变化。
while True:
with tf.Session() as sess:
#注意此处不存在变量初始化的过程
# tf.train.get_checkpoint_state函数会通过checkpoint文件自动
# 找到目录中最新模型的文件名。
ckpt = tf.train.get_checkpoint_state(
train.MODEL_SAVE_PATH)
if ckpt and ckpt.model_checkpoint_path:
# 加载模型。
saver.restore(sess,
ckpt.model_checkpoint_path)
#通过文件名得到模型保存时迭代的轮数,split('/')[-1]即以'/'作为分隔符,留下右边的部分。
global_step = ckpt.model_checkpoint_path\
.split('/')[-1].split('-')[-1]
accuracy_score = sess.run(accuracy,feed_dict=validate_feed)
print("After %s training step(s), validation "
"accuracy = %g" % (global_step, accuracy_score))
else:
print('No checkpoint file found')
return
time.sleep(EVAL_INTERVAL_SECS)
def main(argv=None):
mnist = input_data.read_data_sets("C:/path/to/MNIST_data", one_hot=True)
evaluate(mnist)
if __name__ == '__main__':
tf.app.run()
至此整个程序已经完整了接下来便是意义运行各部分了
train.py运行结果【CPU跑了1个多小时】
evaluate.py的运行结果
完!