TensorFlow Serving Manager

A simple easy to use API for interacting with the TensorFlow Serving server to make predictions and load configuration files

Installation

First clone the repository, and then install using the command:

$   pip install .

Usage

Refer to the notebooks for examples of how to use the package.

Functions are provided to interact with both the RESTful API endpoints, as well as the gRPC API endpoints.

Documentation

Source files can be generated using the command:

$   sphinx-apidoc -o source ../tfserving_manager

Documentation can then be built using the Makefile. First, run make html inside the docs/ directory. Then, launch a server inside docs/build/html using the command python -m http.server.