TensorFlow Server with C++ example:
Either run setup.sh or follow these instructions if the script does not work for you
- install tensorflow serving (https://www.tensorflow.org/tfx/serving/setup)
- run server
- with args: rest_api_port=8501, model_name=TinyModel, model_base_path=$(pwd)/model/
- see setup.sh
- run the python script in models to generat a SaveModel object
- requires tensorflow 2.x (pip install -q tensorflow==2.0.0-rc2)
- make and run the example
- requires libcurl (see setup.sh)