Deploying CNN based Tensorflow model using Tensorflow Serving, Flask, Docker and Docker compose

Deploying a Tensorflow InceptionV3 model with Flask for smoker detection in a contenairized environnement

Project Organization

│───app
│   ├───app.py
│   ├───Dockerfile 
│   ├───requirements.txt
│   ├───static
│   └───templates
│
├───data
│   ├───models
│   ├───non_smoker
│   └───smoker
├───data_tf
│   └───smoker-v-non_smoker
│       ├───testing
│       │   ├───non_smoker
│       │   └───smoker
│       └───training
│           ├───non_smoker
│           └───smoker
├───demo
│       app_demo.gif
│       demo_app.gif
│       img.jpg
│
└───smoker_detection
    │   saved_model.pb
    │
    ├───assets
    └───variables
            variables.data-00000-of-00001
            variables.index

Only 3 steps to run !

1. Clone the repository

git clone https://github.com/amine-akrout/Smoker_detection.git

cd Smoker_detection

2. Build and run the Docker Stack (TensorFlow serving and Web App)

docker-compose build .

docker-compose up

3. Open the Web App and try it out !

The App should be running on http://localhost:5000

What the app looks like

References:

• Convolutional Neural Networks in TensorFlow

• https://github.com/rexsimiloluwah/fastapi-tensorflow-serving-old

• https://github.com/huyhoang17/tensorflow-serving-docker