技巧：使用freeze_graph将SavedModel生成的模型合并成一个pb文件

在训练mnist模型的时候，增加一个参数，可以在训练完毕后导出SavedModel

python mnist.py --export_dir /tmp/mnist_saved_model

训练完成后，会以以下文件结构在/tmp/mnist_saved_model生成模型文件

── 1530159051
   ├── saved_model.pb
   └── variables
       ├── variables.data-00000-of-00001
       └── variables.index

saved_model.pb 保存的是计算图信息
variables 保存的是计算中的变量信息

我的目是想要在Android上运行测试这个模型，但是从官方给的Demo中可以看到，所加载的模型都是单个文件，暂时也没看到加载SavedModel的接口

为了方便Android平台使用，用freeze_graph脚本将SavedModel合并

查看这个脚本的源码，在最开头可以看到

Converts checkpoint variables into Const ops in a standalone GraphDef file.
This script is designed to take a GraphDef proto, a SaverDef proto, and a set of
variable values stored in a checkpoint file, and output a GraphDef with all of
the variable ops converted into const ops containing the values of the
variables.
It's useful to do this when we need to load a single file in C++, especially in
environments like mobile or embedded where we may not have access to the
RestoreTensor ops and file loading calls that they rely on.

这个脚本就是用来将变量转换为常量合并到到计算图中的

具体怎么使用呢？

脚本中给出的例子是这样的：

bazel build tensorflow/python/tools:freeze_graph && \
bazel-bin/tensorflow/python/tools/freeze_graph \
--input_graph=some_graph_def.pb \
--input_checkpoint=model.ckpt-8361242 \
--output_graph=/tmp/frozen_graph.pb --output_node_names=softmax

因为我电脑是完整安装了tensorflow的，所以可以直接使用freeze_graph而不用通过bazel去创建freeze_graph

freeze_graph --input_graph=saved_model.pb --input_checkpoint=variables/variables --output_graph=merge_graph.pb --output_node_names=softmax

执行，然后挂了

  from ._conv import register_converters as _register_converters
Traceback (most recent call last):
  File "/home/xxx/anaconda2/bin/freeze_graph", line 11, in <module>
    sys.exit(run_main())
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/tensorflow/python/tools/freeze_graph.py", line 379, in run_main
    app.run(main=my_main, argv=[sys.argv[0]] + unparsed)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/tensorflow/python/platform/app.py", line 125, in run
    _sys.exit(main(argv))
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/tensorflow/python/tools/freeze_graph.py", line 378, in <lambda>
    my_main = lambda unused_args: main(unused_args, flags)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/tensorflow/python/tools/freeze_graph.py", line 272, in main
    flags.saved_model_tags, checkpoint_version)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/tensorflow/python/tools/freeze_graph.py", line 231, in freeze_graph
    input_graph_def = _parse_input_graph_proto(input_graph, input_binary)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/tensorflow/python/tools/freeze_graph.py", line 174, in _parse_input_graph_proto
    text_format.Merge(f.read(), input_graph_def)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/google/protobuf/text_format.py", line 536, in Merge
    descriptor_pool=descriptor_pool)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/google/protobuf/text_format.py", line 590, in MergeLines
    return parser.MergeLines(lines, message)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/google/protobuf/text_format.py", line 623, in MergeLines
    self._ParseOrMerge(lines, message)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/google/protobuf/text_format.py", line 638, in _ParseOrMerge
    self._MergeField(tokenizer, message)
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/google/protobuf/text_format.py", line 706, in _MergeField
    name = tokenizer.ConsumeIdentifierOrNumber()
  File "/home/xxx/anaconda2/lib/python2.7/site-packages/google/protobuf/text_format.py", line 1166, in ConsumeIdentifierOrNumber
    raise self.ParseError('Expected identifier or number, got %s.' % result)
google.protobuf.text_format.ParseError: 1:1 : Expected identifier or number, got.

什么鬼错，Expected identifier or number 字面意思应该是传入的文件少了点什么东西

再看看freeze_graph脚本的源码

在run_main()函数里看到了一个参数

  parser.add_argument(
      "--input_saved_model_dir",
      type=str,
      default="",
      help="Path to the dir with TensorFlow \'SavedModel\' file and variables.")

原来是用错参数了，噗...

修改一下命令

freeze_graph \
    --input_saved_model_dir=1530159051 \
    --output_node_names=Softmax,ArgMax \
    --output_graph=merge1_graph.pb

再执行，屁都没冒一个，说明执行成功了，没报错，哈哈

看下指定的输出文件有没有生成

ww3:~/source/TFmodels_out/mnist_saved_model$ ls -l
总用量 12804
drwxr-xr-x 3 ww3 ww3     4096  6月 28 12:10 1530159051
-rw-rw-r-- 1 ww3 ww3 13105521  6月 28 17:58 merge1_graph.pb

嗯...具体这个文件能不能直接用还有待研究