我正在尝试从Keras部署到opencv c ++。
我使用mnist数据集训练了一个简单的CNN（我的示例是经过修改的Keras示例）。训练后，我从Keras后端公开了张量流图，并保存了模型和图。
tensorFlowSession = K.get_session()
tf.saved_model.simple_save(tensorFlowSession, newpath + "/TensorFlow", inputs={"x": x}, outputs={"y": y})
tf.train.write_graph(tensorFlowSession.graph_def,newpath + "/TensorFlow", "trainGraph_def.pbtxt")
然后，我尝试在python中使用opencv加载保存的模型，我从python中的opencv开始，但是在c ++中使用opencv时遇到了类似的错误。
net = cv.dnn.readNet(newpath + '/TensorFlow/' + 'saved_model.pb', newpath + '/TensorFlow/' + 'trainGraph.pbtxt')
问题是opencv无法加载tensorflow图，我得到一个错误-
[libprotobuf错误/io/opencv/3rdparty/protobuf/src/google/protobuf/wire_format_lite.cc:629]解析协议缓冲区时，字符串字段'tensorflow.FunctionDef.Node.ret'包含无效的UTF-8数据。如果要发送原始字节，请使用“字节”类型。
Saving and loading a tensorflow graph using opencv应该很简单，我在这里想念的是什么？参见附件中的代码。
'''Trains a simple convnet on the MNIST dataset.
Gets to 99.25% test accuracy after 12 epochs
(there is still a lot of margin for parameter tuning).
16 seconds per epoch on a GRID K520 GPU.
'''
from __future__ import print_function
import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras import backend as K
import tensorflow as tf
import datetime
import cv2 as cv
from pathlib import Path
import numpy as np
from os import listdir
from os.path import isfile, join
import os
os.environ['CUDA_VISIBLE_DEVICES'] = '-1'
batch_size = 128
num_classes = 10
epochs = 1
# input image dimensions
img_rows, img_cols = 28, 28
# the data, split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()
if K.image_data_format() == 'channels_first':
x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
input_shape = (1, img_rows, img_cols)
else:
x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)
x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)
input_shape = (img_rows, img_cols, 1)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
print('x_train shape:', x_train.shape)
print(x_train.shape[0], 'train samples')
print(x_test.shape[0], 'test samples')
# convert class vectors to binary class matrices
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3),
activation='relu',
input_shape=input_shape))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes, activation='softmax'))
model.compile(loss=keras.losses.categorical_crossentropy,
optimizer=keras.optimizers.Adadelta(),
metrics=['accuracy'])
model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_data=(x_test, y_test))
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
# Save as tensor flow graph
# Save the model to pd file
exportModelDir = '/home/iaiai/Documents/Shahar/DataSets/TrainedNetworks/MnistKeras/'
newpath = exportModelDir + str(datetime.datetime.now())
print ("output dir", newpath)
if not os.path.exists(newpath):
os.makedirs(newpath)
x = tf.placeholder(dtype=tf.float32, shape=input_shape)
y = tf.placeholder(dtype=tf.float32, shape=num_classes)
tensorFlowSession = K.get_session()
tf.saved_model.simple_save(tensorFlowSession, newpath + "/TensorFlow", inputs={"x": x}, outputs={"y": y})
# with tf.Session() as sess:
# tf.train.Saver(tf.trainable_variables()).save(sess, newpath + "/TensorFlow/" + 'saveTrainableVariables')
if not os.path.exists(newpath + "/TensorFlow/" + 'saveTrainableVariables/'):
os.makedirs(newpath + "/TensorFlow/" + 'saveTrainableVariables/')
tf.train.Saver(tf.trainable_variables()).save(tensorFlowSession, newpath + "/TensorFlow/saveTrainableVariables/" + 'saveTrainableVariables')
# Write graph in tensorflow format
tf.train.write_graph(tensorFlowSession.graph_def,newpath + "/TensorFlow", "trainGraph_def.pbtxt")
tf.train.write_graph(tensorFlowSession.graph,newpath + "/TensorFlow", "trainGraph.pbtxt")
tf.train.write_graph(tensorFlowSession.graph_def,newpath + "/TensorFlow", "trainGraph_def_notAsText.pbtxt", False)
# now try to load the tensorflow graph from opencv python module
net = cv.dnn.readNet(newpath + '/TensorFlow/' + 'saved_model.pb', newpath + '/TensorFlow/' + 'trainGraph.pbtxt')
这是完整的错误消息
[libprotobuf ERROR /io/opencv/3rdparty/protobuf/src/google/protobuf/wire_format_lite.cc:629] String field 'tensorflow.FunctionDef.Node.ret' contains invalid UTF-8 data when parsing a protocol buffer. Use the 'bytes' type if you intend to send raw bytes.
Traceback (most recent call last):
File "/snap/pycharm-community/79/helpers/pydev/pydevd.py", line 1664, in <module>
main()
File "/snap/pycharm-community/79/helpers/pydev/pydevd.py", line 1658, in main
globals = debugger.run(setup['file'], None, None, is_module)
File "/snap/pycharm-community/79/helpers/pydev/pydevd.py", line 1068, in run
pydev_imports.execfile(file, globals, locals) # execute the script
File "/snap/pycharm-community/79/helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "/home/iaiai/Documents/Shahar/Project2/MnistKerasExport/mnist_cnn.py", line 106, in <module>
net = cv.dnn.readNet(newpath + '/TensorFlow/' + 'saved_model.pb', newpath + '/TensorFlow/' + 'trainGraph.pbtxt')
cv2.error: OpenCV(3.4.2) /io/opencv/modules/dnn/src/tensorflow/tf_io.cpp:44: error: (-2:Unspecified error) FAILED: ReadProtoFromBinaryFile(param_file, param). Failed to parse GraphDef file: /home/iaiai/Documents/Shahar/DataSets/TrainedNetworks/MnistKeras/2018-09-06 08:35:34.025216/TensorFlow/saved_model.pb in function 'ReadTFNetParamsFromBinaryFileOrDie'
答案 0 :(得分：0)
与我的网络相同，但带有* .pbtxt文件。
FAILED: ReadProtoFromTextFile(param_file, param). Failed to parse GraphDef file: ./output_graph.pbtxt" char *
*。pb似乎可以，下面是生成它的代码：
output_nodes = [n.name for n in tf.get_default_graph().as_graph_def().node]
# Freeze the graph
frozen_gdef = tf.graph_util.convert_variables_to_constants(
sess,
sess.graph_def,
output_nodes)
# Save the frozen graph
with open('./model/output_graph.pb', 'wb') as f:
f.write(frozen_gdef.SerializeToString())
希望有帮助！
让我知道您的* .pbtxt是否可以