Created by: iceriver97

我将模型 freeze 之后，使用 inference_model_to_serving 将模型保存为 Serving模型； 但我想知道客户端调用 client.predict()函数时，fetch参数应该是什么？ 不应该是我在freeze模型时使用函数 save_inference_model 时指定的 fetch吗？我将其作为参数得到报错信息：

aistudio@jupyter-380628-755201:~$ python ./client.py 983884_sat.jpg 
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/sklearn/externals/joblib/externals/cloudpickle/cloudpickle.py:47: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
  import imp
WARNING: Logging before InitGoogleLogging() is written to STDERR
I0907 16:32:14.215747  2107 naming_service_thread.cpp:209] brpc::policy::ListNamingService("127.0.0.1:9292"): added 1
Traceback (most recent call last):
  File "./client.py", line 18, in <module>
    fetch_map = client.predict(feed={"image": im}, fetch=["transpose_1.tmp_0"])
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle_serving_client/__init__.py", line 274, in predict
    "Fetch names should not be empty or out of saved fetch list.")
ValueError: Fetch names should not be empty or out of saved fetch list.

我无法打开 serving_client_conf.prototxt 文件，是不是在该文件中可以得到我想要的信息呢？ 这里是该文件

serving_client_conf.zip 固化模型的代码：

fluid.io.save_inference_model(
        cfg.FREEZE.SAVE_DIR,
        feeded_var_names=[image.name],
        target_vars=[logit_out],
        executor=exe,
        main_program=infer_prog,
        model_filename=cfg.FREEZE.MODEL_FILENAME,
        params_filename=cfg.FREEZE.PARAMS_FILENAME)

我是不是应该指定client.predict()> fetch = 这里 logit_out.name?