最近需要將使用keras訓練的模型移植到手機上使用， 因此需要轉換到tensorflow的二進位制模型。

折騰一下午，終於找到一個合適的方法，廢話不多說，直接上程式碼：

# coding=utf-8
import sys

from keras.models import load_model
import tensorflow as tf
import os
import os.path as osp
from keras import backend as K


def freeze_session(session, keep_var_names=None, output_names=None
 
, clear_devices=True):
    """
    Freezes the state of a session into a prunned computation graph.
    Creates a new computation graph where variable nodes are replaced by
    constants taking their current value in the session. The new graph will be
    prunned so subgraphs that are not neccesary to compute the requested
 
    outputs are removed.
    @param session The TensorFlow session to be frozen.
    @param keep_var_names A list of variable names that should not be frozen,
                          or None to freeze all the variables in the graph.
    @param output_names Names of the relevant graph outputs.
    @param clear_devices Remove the device directives from the graph for better portability.
 
    @return The frozen graph definition.
    """
from tensorflow.python.framework.graph_util import convert_variables_to_constants
    graph = session.graph
    with graph.as_default():
        freeze_var_names = list(set(v.op.name for v in tf.global_variables()).difference(keep_var_names or []))
        output_names = output_names or []
        output_names += [v.op.name for v in tf.global_variables()]
        input_graph_def = graph.as_graph_def()
        if clear_devices:
            for node in input_graph_def.node:
                node.device = ""
frozen_graph = convert_variables_to_constants(session, input_graph_def,
output_names, freeze_var_names)
        return frozen_graph


input_fld = sys.path[0]
weight_file = 'your_model.h5'
output_graph_name = 'tensor_model.pb'
output_fld = input_fld + '/tensorflow_model/'
if not os.path.isdir(output_fld):
    os.mkdir(output_fld)
weight_file_path = osp.join(input_fld, weight_file)

K.set_learning_phase(0)
net_model = load_model(weight_file_path)


print('input is :', net_model.input.name)
print ('output is:', net_model.output.name)

sess = K.get_session()

frozen_graph = freeze_session(K.get_session(), output_names=[net_model.output.op.name])

from tensorflow.python.framework import graph_io

graph_io.write_graph(frozen_graph, output_fld, output_graph_name, as_text=False)

print('saved the constant graph (ready for inference) at: ', osp.join(output_fld, output_graph_name))

上面程式碼實現儲存到當前目錄的tensor_model目錄下。

驗證：

import tensorflow as tf
import numpy as np
import PIL.Image as Image
import cv2


def recognize(jpg_path, pb_file_path):
    with tf.Graph().as_default():
        output_graph_def = tf.GraphDef()

        with open(pb_file_path, "rb") as f:
            output_graph_def.ParseFromString(f.read())
            tensors = tf.import_graph_def(output_graph_def, name="")
            print tensors

        with tf.Session() as sess:
            init = tf.global_variables_initializer()
            sess.run(init)

            op = sess.graph.get_operations()

          
            for m in op:
                print(m.values())

            input_x = sess.graph.get_tensor_by_name("convolution2d_1_input:0")  #具體名稱看上一段程式碼的input.name
            print input_x

            out_softmax = sess.graph.get_tensor_by_name("activation_4/Softmax:0") #具體名稱看上一段程式碼的output.name

            print out_softmax

            img = cv2.imread(jpg_path, 0)
            img_out_softmax = sess.run(out_softmax,
feed_dict={input_x: 1.0 - np.array(img).reshape((-1,28, 28, 1)) / 255.0})

            print "img_out_softmax:", img_out_softmax
            prediction_labels = np.argmax(img_out_softmax, axis=1)
            print "label:", prediction_labels


pb_path = 'tensorflow_model/constant_graph_weights.pb'
img = 'test/6/8_48.jpg'
recognize(img, pb_path)

參考：

https://github.com/amir-abdi/keras_to_tensorflow