如何在TF网站中设置Tensorflow Serving？

Question

I am trying to learn and setup TF serving as described here

I trained my model using retrain.py file and all necessary model files are in "saved_models" folder

Now, I want to update the server command

    # Start TensorFlow Serving container and open the REST API port
docker run -t --rm -p 8501:8501 \
   -v "$TESTDATA/saved_model_half_plus_two_cpu:/models/half_plus_two" \
   -e MODEL_NAME=half_plus_two \
   tensorflow/serving &

I have a hard time understanding to change the example path to my current setup. I don't understand how "/models/half_plus_two" got there since it's not available in the example folders.

This is what I did here

docker run -t --rm -p 8501:8501 -v "c:/tmp/saved_models:/models/1" -e MODEL_NAME=c:\tmp\saved_models\1\saved_model.pb

But it fails

C:\tmp\serving>docker run -t --rm -p 8501:8501 -v "c:/tmp/saved_models:/models/1" -e MODEL_NAME=c:\tmp\saved_models\1\saved_model.pb
"docker run" requires at least 1 argument.
See 'docker run --help'.

Usage:  docker run [OPTIONS] IMAGE [COMMAND] [ARG...]

Run a command in a new container

Any help is really appreciated.

Answer 1

Did you docker pull the tensorflow/serving image? Also, you need to provide docker run an "image"(the one that you pulled before).

docker pull tensorflow/serving