hunjianghu/gzy/tesorflow/改进后的训练模型.py

# -*- coding=UTF-8 -*-
import os
import tensorflow as tf
import numpy as np
import time
from PIL import Image

w = 64
h = 64
c = 3
TOTAL_TYPE = 5

path = 'D:/tensorflow/imgaes/'
model_path='D:/tensorflow/saver/model.ckpt'

def read_source():
    imgs,labels = [],[]
    for dir in os.listdir(path):
        idx = int(dir)
        folder = os.path.join(path,dir)
        zeros = np.zeros(TOTAL_TYPE)
        zeros[idx]=1
        print("folder :%s"%(folder))
        total = 0
        for f in os.listdir(folder):
            file = os.path.join(folder,f)
            image = Image.open(file).convert('RGB').resize((w, h), Image.ANTIALIAS)
            arr = np.asarray(image)
            imgs.append(arr)
            labels.append(zeros)
            total += 1
        print(total)
    return np.asarray(imgs,np.float32),np.asarray(labels,np.int32)

data,label=read_source()
num_example = data.shape[0]
arr = np.arange(num_example)
np.random.shuffle(arr)
data = data[arr]
label = label[arr]

# 将所有数据分为训练集和验证集
ratio = 0.8
s = np.int(num_example * ratio)
x_train = data[:s]
y_train = label[:s]
x_val = data[s:]
y_val = label[s:]


# -----------------构建网络----------------------
x = tf.placeholder(tf.float32, shape=[None, w, h, 3], name='x')
y_ = tf.placeholder(tf.float32, shape=[None,5 ], name='y_')


def inference(input_tensor, train, regularizer):
    with tf.variable_scope('layer1-conv1'):
        conv1_weights = tf.get_variable("weight", [5, 5, 3, 64],
                                        initializer=tf.truncated_normal_initializer(stddev=0.1))
        conv1_biases = tf.get_variable("bias", [64], 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))
        relu1 = tf.layers.batch_normalization(relu1,training=train)

    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", [3, 3, 64, 128],
                                        initializer=tf.truncated_normal_initializer(stddev=0.1))
        conv2_biases = tf.get_variable("bias", [128], 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))
        relu2 = tf.layers.batch_normalization(relu2, training=train)

    with tf.name_scope("layer4-pool2"):
        pool2 = tf.nn.max_pool(relu2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')

    with tf.variable_scope("layer5-conv3"):
        conv3_weights = tf.get_variable("weight", [3, 3, 128, 256],
                                        initializer=tf.truncated_normal_initializer(stddev=0.1))
        conv3_biases = tf.get_variable("bias", [256], initializer=tf.constant_initializer(0.0))
        conv3 = tf.nn.conv2d(pool2, conv3_weights, strides=[1, 1, 1, 1], padding='SAME')
        relu3 = tf.nn.relu(tf.nn.bias_add(conv3, conv3_biases))
        relu3 = tf.layers.batch_normalization(relu3, training=train)


    with tf.name_scope("layer8-pool4"):
        pool4 = tf.nn.max_pool(relu3, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
        norm4 = tf.nn.lrn(pool4, 4, bias=1.0, alpha=0.001 / 9.0, beta=0.75)
        nodes = 8 * 8* 256
        reshaped = tf.reshape(norm4, [-1, nodes])

    with tf.variable_scope('layer9-fc1'):
        fc1_weights = tf.get_variable("weight", [nodes, 1024],
                                      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", [1024], initializer=tf.constant_initializer(0.1))

        fc1 = tf.nn.relu(tf.matmul(reshaped, fc1_weights) + fc1_biases)
        if train: fc1 = tf.nn.dropout(fc1, 0.8)

    with tf.variable_scope('layer10-fc2'):
        fc2_weights = tf.get_variable("weight", [1024, 512],
                                      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", [512], initializer=tf.constant_initializer(0.1))

        fc2 = tf.nn.relu(tf.matmul(fc1, fc2_weights) + fc2_biases)
        if train: fc2 = tf.nn.dropout(fc2, 0.8)

    with tf.variable_scope('layer11-fc3'):
        fc3_weights = tf.get_variable("weight", [512, 5],
                                      initializer=tf.truncated_normal_initializer(stddev=0.1))
        if regularizer != None: tf.add_to_collection('losses', regularizer(fc3_weights))
        fc3_biases = tf.get_variable("bias", [5], initializer=tf.constant_initializer(0.1))
        logit = tf.matmul(fc2, fc3_weights) + fc3_biases

    return logit


# ---------------------------网络结束---------------------------
regularizer = tf.contrib.layers.l2_regularizer(0.0001)
logits = inference(x, False, regularizer)

# (小处理)将logits乘以1赋值给logits_eval，定义name，方便在后续调用模型时通过tensor名字调用输出tensor
b = tf.constant(value=1, dtype=tf.float32)
logits_eval = tf.multiply(logits, b, name='logits_eval')

# loss = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits, labels=y_)
# loss = -tf.reduce_sum(y_*tf.log(logits))
loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(logits=logits,labels=y_))
update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
with tf.control_dependencies(update_ops):
    train_op = tf.train.AdamOptimizer(learning_rate=0.001).minimize(loss)
correct_prediction = tf.equal(tf.argmax(logits, 1),tf.argmax(y_,1))
acc = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))


# 定义一个函数，按批次取数据
def minibatches(inputs=None, targets=None, batch_size=None, shuffle=False):
    assert len(inputs) == len(targets)
    if shuffle:
        indices = np.arange(len(inputs))
        np.random.shuffle(indices)
    for start_idx in range(0, len(inputs) - batch_size + 1, batch_size):
        if shuffle:
            excerpt = indices[start_idx:start_idx + batch_size]
        else:
            excerpt = slice(start_idx, start_idx + batch_size)
        yield inputs[excerpt], targets[excerpt]


# 训练和测试数据，可将n_epoch设置更大一些


n_epoch = 200
batch_size = 64
saver = tf.train.Saver()
sess = tf.Session()
sess.run(tf.global_variables_initializer())
for epoch in range(n_epoch):
    start_time = time.time()

    # training
    train_loss, train_acc, n_batch = 0, 0, 0
    for x_train_a, y_train_a in minibatches(x_train, y_train, batch_size, shuffle=True):
        _, err, ac = sess.run([train_op, loss, acc], feed_dict={x: x_train_a, y_: y_train_a})
        train_loss += err;
        train_acc += ac;
        n_batch += 1
    print("   train loss: %f" % (np.sum(train_loss) / n_batch))
    print("   train acc: %f" % (np.sum(train_acc) / n_batch))

    # validation
    val_loss, val_acc, n_batch = 0, 0, 0
    for x_val_a, y_val_a in minibatches(x_val, y_val, batch_size, shuffle=False):
        err, ac = sess.run([loss, acc], feed_dict={x: x_val_a, y_: y_val_a})
        val_loss += err
        val_acc += ac
        n_batch += 1
    print("   validation loss: %f" % (np.sum(val_loss) / n_batch))
    print("   validation acc: %f" % (np.sum(val_acc) / n_batch))
saver.save(sess,model_path)
sess.close()