Question

我是张力流的初学者。我想量化已经训练过的初始模型。我按照步骤运行以下代码：

python -m tensorflow.python.tools.optimize_for_inference \
  --input=tf_files/retrained_graph.pb \
  --output=tf_files/optimized_graph.pb \
  --input_names="input" \
  --output_names="final_result"

但我收到错误：

Traceback (most recent call last):
  File "C:\Users\Student\Anaconda3\lib\runpy.py", line 193, in _run_module_as_ma
in
    "__main__", mod_spec)
  File "C:\Users\Student\Anaconda3\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Users\Student\Anaconda3\lib\site-packages\tensorflow\python\tools\opt
imize_for_inference.py", line 146, in <module>
    app.run(main=main, argv=[sys.argv[0]] + unparsed)
  File "C:\Users\Student\Anaconda3\lib\site-packages\tensorflow\python\platform\
app.py", line 126, in run
    _sys.exit(main(argv))
  File "C:\Users\Student\Anaconda3\lib\site-packages\tensorflow\python\tools\opt
imize_for_inference.py", line 90, in main
    FLAGS.output_names.split(","), FLAGS.placeholder_type_enum)
  File "C:\Users\Student\Anaconda3\lib\site-packages\tensorflow\python\tools\opt
imize_for_inference_lib.py", line 109, in optimize_for_inference
    placeholder_type_enum)
  File "C:\Users\Student\Anaconda3\lib\site-packages\tensorflow\python\tools\str
ip_unused_lib.py", line 83, in strip_unused
    raise KeyError("The following input nodes were not found: %s\n" % not_found)

KeyError: "The following input nodes were not found: {'input'}\n"

请有人帮我解决此问题。

Answer 1

更改

--input_names="input"

到

--input_names="Mul"

或者如果您在Windows上，只需在终端中从当前工作区运行以下内容

python -m tensorflow.python.tools.optimize_for_inference --input 
retrained_graph.pb --output optimized_graph.pb --input_names="Mul" -- 
output_names="final_result"

无法优化训练的张量流图

1 个答案: