/tensorflow-deployment

ML deployment boilerplate using Tensorflow Serving with Docker and Flask

Primary LanguagePython

Deploying a Custom Deep Learning Tensorflow Model using Tensorflow Serving with Docker Compose and Flask

Environment Prerequisites

Make sure you have Docker Compose installed in your computer.

Execution

  1. Clone the repository into your local machine using:

    git clone https://github.com/nardienapratama/tensorflow-deployment.git

  2. Enter the tensorflow-deployment directory.

  3. Start up the application by running docker-compose up --build or sudo docker-compose up --build if you are in Linux. After having run the command, assuming you have not deleted the container, you can run the application again without using the --build tag, i.e. docker-compose up or sudo docker-compose up.

  4. Enter http://localhost:5000/ in your browser to see the application running.

Sources Used

This boilerplate is based on this tutorial and the image classification model was created based on this tutorial, though modifications have been made accordingly.