tensorflow將訓練好的模型freeze,即將權重固化到圖裏面,並使用該模型進行預測
ML主要分爲訓練和預測兩個階段,此教程就是將訓練好的模型freeze並保存下來.freeze的含義就是將該模型的圖結構和該模型的權重固化到一塊兒了.也即加載freeze的模型以後,馬上可以使用了。node
下面使用一個簡單的demo來詳細解釋該過程,python
1、首先運行腳本tiny_model.py網絡
#-*- coding:utf-8 -*-
import tensorflow as tf
import numpy as np
with tf.variable_scope('Placeholder'):
inputs_placeholder = tf.placeholder(tf.float32, name='inputs_placeholder', shape=[None, 10])
labels_placeholder = tf.placeholder(tf.float32, name='labels_placeholder', shape=[None, 1])
with tf.variable_scope('NN'):
W1 = tf.get_variable('W1', shape=[10, 1], initializer=tf.random_normal_initializer(stddev=1e-1))
b1 = tf.get_variable('b1', shape=[1], initializer=tf.constant_initializer(0.1))
W2 = tf.get_variable('W2', shape=[10, 1], initializer=tf.random_normal_initializer(stddev=1e-1))
b2 = tf.get_variable('b2', shape=[1], initializer=tf.constant_initializer(0.1))
a = tf.nn.relu(tf.matmul(inputs_placeholder, W1) + b1)
a2 = tf.nn.relu(tf.matmul(inputs_placeholder, W2) + b2)
y = tf.div(tf.add(a, a2), 2)
with tf.variable_scope('Loss'):
loss = tf.reduce_sum(tf.square(y - labels_placeholder) / 2)
with tf.variable_scope('Accuracy'):
predictions = tf.greater(y, 0.5, name="predictions")
correct_predictions = tf.equal(predictions, tf.cast(labels_placeholder, tf.bool), name="correct_predictions")
accuracy = tf.reduce_mean(tf.cast(correct_predictions, tf.float32))
adam = tf.train.AdamOptimizer(learning_rate=1e-3)
train_op = adam.minimize(loss)
# generate_data
inputs = np.random.choice(10, size=[10000, 10])
labels = (np.sum(inputs, axis=1) > 45).reshape(-1, 1).astype(np.float32)
print('inputs.shape:', inputs.shape)
print('labels.shape:', labels.shape)
test_inputs = np.random.choice(10, size=[100, 10])
test_labels = (np.sum(test_inputs, axis=1) > 45).reshape(-1, 1).astype(np.float32)
print('test_inputs.shape:', test_inputs.shape)
print('test_labels.shape:', test_labels.shape)
batch_size = 32
epochs = 10
batches = []
print("%d items in batch of %d gives us %d full batches and %d batches of %d items" % (
len(inputs),
batch_size,
len(inputs) // batch_size,
batch_size - len(inputs) // batch_size,
len(inputs) - (len(inputs) // batch_size) * 32)
)
for i in range(len(inputs) // batch_size):
batch = [ inputs[batch_size*i:batch_size*i+batch_size], labels[batch_size*i:batch_size*i+batch_size] ]
batches.append(list(batch))
if (i + 1) * batch_size < len(inputs):
batch = [ inputs[batch_size*(i + 1):],labels[batch_size*(i + 1):] ]
batches.append(list(batch))
print("Number of batches: %d" % len(batches))
print("Size of full batch: %d" % len(batches[0]))
print("Size if final batch: %d" % len(batches[-1]))
global_count = 0
with tf.Session() as sess:
#sv = tf.train.Supervisor()
#with sv.managed_session() as sess:
sess.run(tf.initialize_all_variables())
for i in range(epochs):
for batch in batches:
# print(batch[0].shape, batch[1].shape)
train_loss , _= sess.run([loss, train_op], feed_dict={
inputs_placeholder: batch[0],
labels_placeholder: batch[1]
})
# print('train_loss: %d' % train_loss)
if global_count % 100 == 0:
acc = sess.run(accuracy, feed_dict={
inputs_placeholder: test_inputs,
labels_placeholder: test_labels
})
print('accuracy: %f' % acc)
global_count += 1
acc = sess.run(accuracy, feed_dict={
inputs_placeholder: test_inputs,
labels_placeholder: test_labels
})
print("final accuracy: %f" % acc)
#在session當中就要將模型進行保存
saver = tf.train.Saver()
last_chkp = saver.save(sess, 'results/graph.chkp')
#sv.saver.save(sess, 'results/graph.chkp')
for op in tf.get_default_graph().get_operations():
print(op.name)
說明：saver.save必須在session裏面,由於在session裏面,整個圖纔是激活的,纔可以將參數存進來,使用save以後可以獲得以下的文件：session
說明：app
.data:存放的是權重參數dom
.meta:存放的是圖和metadata,metadata是其餘配置的數據ui
若是想將咱們的模型固化，讓別人可以使用，咱們僅僅須要的是圖和參數，metadata是不須要的this
2、綜合上述幾個文件,生成可使用的模型的步驟以下：rest
一、恢復咱們保存的圖code
二、開啓一個Session，而後載入該圖要求的權重
三、刪除對預測無關的metadata
四、將處理好的模型序列化以後保存
運行freeze.py
#-*- coding:utf-8 -*-
import os, argparse
import tensorflow as tf
from tensorflow.python.framework import graph_util
dir = os.path.dirname(os.path.realpath(__file__))
def freeze_graph(model_folder):
# We retrieve our checkpoint fullpath
checkpoint = tf.train.get_checkpoint_state(model_folder)
input_checkpoint = checkpoint.model_checkpoint_path
# We precise the file fullname of our freezed graph
absolute_model_folder = "/".join(input_checkpoint.split('/')[:-1])
output_graph = absolute_model_folder + "/frozen_model.pb"
# Before exporting our graph, we need to precise what is our output node
# this variables is plural, because you can have multiple output nodes
#freeze以前必須明確哪一個是輸出結點,也就是咱們要獲得推論結果的結點
#輸出結點能夠看咱們模型的定義
#只有定義了輸出結點,freeze纔會把獲得輸出結點所必要的結點都保存下來,或者哪些結點能夠丟棄
#因此,output_node_names必須根據不一樣的網絡進行修改
output_node_names = "Accuracy/predictions"
# We clear the devices, to allow TensorFlow to control on the loading where it wants operations to be calculated
clear_devices = True
# We import the meta graph and retrive a Saver
saver = tf.train.import_meta_graph(input_checkpoint + '.meta', clear_devices=clear_devices)
# We retrieve the protobuf graph definition
graph = tf.get_default_graph()
input_graph_def = graph.as_graph_def()
#We start a session and restore the graph weights
#這邊已經將訓練好的參數加載進來,也即最後保存的模型是有圖,而且圖裏面已經有參數了,因此才叫作是frozen
#至關於將參數已經固化在了圖當中
with tf.Session() as sess:
saver.restore(sess, input_checkpoint)
# We use a built-in TF helper to export variables to constant
output_graph_def = graph_util.convert_variables_to_constants(
sess,
input_graph_def,
output_node_names.split(",") # We split on comma for convenience
)
# Finally we serialize and dump the output graph to the filesystem
with tf.gfile.GFile(output_graph, "wb") as f:
f.write(output_graph_def.SerializeToString())
print("%d ops in the final graph." % len(output_graph_def.node))
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument("--model_folder", type=str, help="Model folder to export")
args = parser.parse_args()
freeze_graph(args.model_folder)
說明：對於freeze操做,咱們須要定義輸出結點的名字.由於網絡實際上是比較複雜的,定義了輸出結點的名字,那麼freeze的時候就只把輸出該結點所須要的子圖都固化下來,其餘無關的就捨棄掉.由於咱們freeze模型的目的是接下來作預測.因此,通常狀況下,output_node_names就是咱們預測的目標.
3、加載freeze後的模型,注意該模型已是包含圖和相應的參數了.因此,咱們不須要再加載參數進來.也即該模型加載進來已是可使用了.
#-*- coding:utf-8 -*-
import argparse
import tensorflow as tf
def load_graph(frozen_graph_filename):
# We parse the graph_def file
with tf.gfile.GFile(frozen_graph_filename, "rb") as f:
graph_def = tf.GraphDef()
graph_def.ParseFromString(f.read())
# We load the graph_def in the default graph
with tf.Graph().as_default() as graph:
tf.import_graph_def(
graph_def,
input_map=None,
return_elements=None,
name="prefix",
op_dict=None,
producer_op_list=None
)
return graph
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument("--frozen_model_filename", default="results/frozen_model.pb", type=str, help="Frozen model file to import")
args = parser.parse_args()
#加載已經將參數固化後的圖
graph = load_graph(args.frozen_model_filename)
# We can list operations
#op.values() gives you a list of tensors it produces
#op.name gives you the name
#輸入,輸出結點也是operation,因此,咱們能夠獲得operation的名字
for op in graph.get_operations():
print(op.name,op.values())
# prefix/Placeholder/inputs_placeholder
# ...
# prefix/Accuracy/predictions
#操做有:prefix/Placeholder/inputs_placeholder
#操做有:prefix/Accuracy/predictions
#爲了預測,咱們須要找到咱們須要feed的tensor,那麼就須要該tensor的名字
#注意prefix/Placeholder/inputs_placeholder僅僅是操做的名字,prefix/Placeholder/inputs_placeholder:0纔是tensor的名字
x = graph.get_tensor_by_name('prefix/Placeholder/inputs_placeholder:0')
y = graph.get_tensor_by_name('prefix/Accuracy/predictions:0')
with tf.Session(graph=graph) as sess:
y_out = sess.run(y, feed_dict={
x: [[3, 5, 7, 4, 5, 1, 1, 1, 1, 1]] # < 45
})
print(y_out) # [[ 0.]] Yay!
print ("finish")
說明：
一、在預測的過程當中,當把freeze後的模型加載進來後,咱們只須要定義好輸入的tensor和目標tensor便可
二、在這裏要注意一下tensor_name和ops_name,
注意prefix/Placeholder/inputs_placeholder僅僅是操做的名字,prefix/Placeholder/inputs_placeholder:0纔是tensor的名字
x = graph.get_tensor_by_name('prefix/Placeholder/inputs_placeholder:0')必定要使用tensor的名字
三、要獲取圖中ops的名字和對應的tensor的名字,可用以下的代碼
# We can list operations
#op.values() gives you a list of tensors it produces
#op.name gives you the name
#輸入,輸出結點也是operation,因此,咱們能夠獲得operation的名字
for op in graph.get_operations():
print(op.name,op.values())
=============================================================================================================================
上面是使用了Saver()來保存模型,也可使用sv = tf.train.Supervisor()來保存模型
#-*- coding:utf-8 -*-
import tensorflow as tf
import numpy as np
with tf.variable_scope('Placeholder'):
inputs_placeholder = tf.placeholder(tf.float32, name='inputs_placeholder', shape=[None, 10])
labels_placeholder = tf.placeholder(tf.float32, name='labels_placeholder', shape=[None, 1])
with tf.variable_scope('NN'):
W1 = tf.get_variable('W1', shape=[10, 1], initializer=tf.random_normal_initializer(stddev=1e-1))
b1 = tf.get_variable('b1', shape=[1], initializer=tf.constant_initializer(0.1))
W2 = tf.get_variable('W2', shape=[10, 1], initializer=tf.random_normal_initializer(stddev=1e-1))
b2 = tf.get_variable('b2', shape=[1], initializer=tf.constant_initializer(0.1))
a = tf.nn.relu(tf.matmul(inputs_placeholder, W1) + b1)
a2 = tf.nn.relu(tf.matmul(inputs_placeholder, W2) + b2)
y = tf.div(tf.add(a, a2), 2)
with tf.variable_scope('Loss'):
loss = tf.reduce_sum(tf.square(y - labels_placeholder) / 2)
with tf.variable_scope('Accuracy'):
predictions = tf.greater(y, 0.5, name="predictions")
correct_predictions = tf.equal(predictions, tf.cast(labels_placeholder, tf.bool), name="correct_predictions")
accuracy = tf.reduce_mean(tf.cast(correct_predictions, tf.float32))
adam = tf.train.AdamOptimizer(learning_rate=1e-3)
train_op = adam.minimize(loss)
# generate_data
inputs = np.random.choice(10, size=[10000, 10])
labels = (np.sum(inputs, axis=1) > 45).reshape(-1, 1).astype(np.float32)
print('inputs.shape:', inputs.shape)
print('labels.shape:', labels.shape)
test_inputs = np.random.choice(10, size=[100, 10])
test_labels = (np.sum(test_inputs, axis=1) > 45).reshape(-1, 1).astype(np.float32)
print('test_inputs.shape:', test_inputs.shape)
print('test_labels.shape:', test_labels.shape)
batch_size = 32
epochs = 10
batches = []
print("%d items in batch of %d gives us %d full batches and %d batches of %d items" % (
len(inputs),
batch_size,
len(inputs) // batch_size,
batch_size - len(inputs) // batch_size,
len(inputs) - (len(inputs) // batch_size) * 32)
)
for i in range(len(inputs) // batch_size):
batch = [ inputs[batch_size*i:batch_size*i+batch_size], labels[batch_size*i:batch_size*i+batch_size] ]
batches.append(list(batch))
if (i + 1) * batch_size < len(inputs):
batch = [ inputs[batch_size*(i + 1):],labels[batch_size*(i + 1):] ]
batches.append(list(batch))
print("Number of batches: %d" % len(batches))
print("Size of full batch: %d" % len(batches[0]))
print("Size if final batch: %d" % len(batches[-1]))
global_count = 0
#with tf.Session() as sess:
sv = tf.train.Supervisor()
with sv.managed_session() as sess:
#sess.run(tf.initialize_all_variables())
for i in range(epochs):
for batch in batches:
# print(batch[0].shape, batch[1].shape)
train_loss , _= sess.run([loss, train_op], feed_dict={
inputs_placeholder: batch[0],
labels_placeholder: batch[1]
})
# print('train_loss: %d' % train_loss)
if global_count % 100 == 0:
acc = sess.run(accuracy, feed_dict={
inputs_placeholder: test_inputs,
labels_placeholder: test_labels
})
print('accuracy: %f' % acc)
global_count += 1
acc = sess.run(accuracy, feed_dict={
inputs_placeholder: test_inputs,
labels_placeholder: test_labels
})
print("final accuracy: %f" % acc)
#在session當中就要將模型進行保存
#saver = tf.train.Saver()
#last_chkp = saver.save(sess, 'results/graph.chkp')
sv.saver.save(sess, 'results/graph.chkp')
for op in tf.get_default_graph().get_operations():
print(op.name)
注意：使用了sv = tf.train.Supervisor(),就不須要再初始化了,將sess.run(tf.initialize_all_variables())註釋掉,不然會報錯.
相關文章
- 1. 【TensorFlow】將tensorflow訓練好的模型移植到Android
- 2. 將tensorflow訓練好的模型部署成sercice服務,並做預測
- 3. 使用Python訓練迴歸模型並進行預測
- 4. 將TensorFlow訓練好的模型遷移到Android APP上（TensorFlowLite）
- 5. TensorFlow 使用預訓練模型 ResNet-50
- 6. Tensorflow在訓練好的模型上進行測試
- 7. 使用tensorflow進行mnist數字識別【模型訓練+預測+模型保存+模型恢復】
- 8. C++調用tensorflow 訓練好的模型
- 9. 使用Keras預訓練模型ResNet50進行圖像分類
- 10. 如何將TensorFlow訓練的模型移植到Android手機
- 更多相關文章...
- • ASP.NET MVC - 模型 - ASP.NET 教程
- • Scala 函數柯里化(Currying) - Scala教程
- • 委託模式
- • 使用阿里雲OSS+CDN部署前端頁面與加速靜態資源
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
最新文章
- 1. No provider available from registry 127.0.0.1:2181 for service com.ddbuy.ser 解決方法
- 2. Qt5.7以上調用虛擬鍵盤(支持中文)，以及源碼修改(可拖動，水平縮放)
- 3. 軟件測試面試- 購物車功能測試用例設計
- 4. ElasticSearch（概念篇）：你知道的, 爲了搜索…
- 5. redux理解
- 6. gitee創建第一個項目
- 7. 支持向量機之硬間隔（一步步推導，通俗易懂）
- 8. Mysql 異步複製延遲的原因及解決方案
- 9. 如何在運行SEPM配置嚮導時將不可認的複雜數據庫密碼改爲簡單密碼
- 10. windows系統下tftp服務器使用
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. 【TensorFlow】將tensorflow訓練好的模型移植到Android
- 2. 將tensorflow訓練好的模型部署成sercice服務,並做預測
- 3. 使用Python訓練迴歸模型並進行預測
- 4. 將TensorFlow訓練好的模型遷移到Android APP上（TensorFlowLite）
- 5. TensorFlow 使用預訓練模型 ResNet-50
- 6. Tensorflow在訓練好的模型上進行測試
- 7. 使用tensorflow進行mnist數字識別【模型訓練+預測+模型保存+模型恢復】
- 8. C++調用tensorflow 訓練好的模型
- 9. 使用Keras預訓練模型ResNet50進行圖像分類
- 10. 如何將TensorFlow訓練的模型移植到Android手機