重点在于,第一个文件用于 训练,保存图meta和训练好的参数data(后缀),在另一个文件中导入这个图和训练好的参数,用于预测或者接着训练。大大减少了另一个文件里的 重复
1. 第一种情况是,产生变量的代码和恢复变量的代码在同一个文件时,可以直接如下调用:
# 建模型
saver = tf.train.Saver()
with tf.Session() as sess:
# 存模型,注意此处的model是文件名,不是路径
saver.save(sess, "/tmp/model")
with tf.Session() as sess:
# 恢复模型
saver.restore(sess, "/tmp/model")
2.第二种情况,不想在另一个文件中,把产生变量的 一大堆代码重敲一遍,可以直接从保存好的 meta文件和data文件中恢复出来
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2019/9/9 20:49
# @Author : ZZL
# @File : 保存检查点文件,并恢复.py
import tensorflow as tf
# Saving contents and operations.
v1 = tf.placeholder(tf.float32, name="v1")
v2 = tf.placeholder(tf.float32, name="v2")
v3 = tf.multiply(v1, v2)
vx = tf.Variable(10.0, name="vx")
v4 = tf.add(v3, vx, name="v4")
saver = tf.train.Saver([vx])
with tf.Session() as sess:
with tf.device('/cpu:0'):
sess.run(tf.global_variables_initializer())
sess.run(vx.assign(tf.add(vx, vx)))
result = sess.run(v4, feed_dict={v1: 12.0, v2: 3.3})
print(result)
print(saver.save(sess, "./model_ex1")) # 该方法返回新创建的检查点文件的路径前缀。这个字符串可以直接传递给对“restore()”的调用。
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2019/9/9 20:54
# @Author : ZZL
# @File : 恢复文件.py
import tensorflow as tf
saver = tf.train.import_meta_graph("./model_ex1.meta")
sess = tf.Session()
saver.restore(sess, "./model_ex1")
result = sess.run("v4:0", feed_dict={"v1:0": 12.0, "v2:0": 3.3})
print(result)
先来个空图,loaded_graph,在会话中,导入之前构建好的图的文件 后缀meta,loader.restore(sess, save_model_path)
在当前的loaded_graph中,导入构建好的图和图上的变量值。
def test_model():
test_features, test_labels = pickle.load(open('preprocess_test.p', mode='rb'))
loaded_graph = tf.Graph() # <tensorflow.python.framework.ops.Graph object at 0x0000017CB3702320>
# print( loaded_graph)
# print(tf.get_default_graph()) # <tensorflow.python.framework.ops.Graph object at 0x0000017C9A0C0C50>
with tf.Session(graph=loaded_graph) as sess:
# 读取模型
loader = tf.train.import_meta_graph(save_model_path + '.meta')
print(loader)
loader.restore(sess, save_model_path)
print(tf.get_default_graph()) # <tensorflow.python.framework.ops.Graph object at 0x0000017CB3702320>
# 从已经读入的模型中 获取tensors
loaded_x = loaded_graph.get_tensor_by_name('x:0')
loaded_y = loaded_graph.get_tensor_by_name('y:0')
loaded_keep_prob = loaded_graph.get_tensor_by_name('keep_prob:0')
loaded_logits = loaded_graph.get_tensor_by_name('logits:0')
loaded_acc = loaded_graph.get_tensor_by_name('accuracy:0')
# 获取每个batch的准确率,再求平均值,这样可以节约内存
test_batch_acc_total = 0
test_batch_count = 0
for test_feature_batch, test_label_batch in helper.batch_features_labels(test_features, test_labels, batch_size):
test_batch_acc_total += sess.run(
loaded_acc,
feed_dict={loaded_x: test_feature_batch, loaded_y: test_label_batch, loaded_keep_prob: 1.0})
test_batch_count += 1
参考: