Skip to content

Latest commit

 

History

History
45 lines (39 loc) · 1.83 KB

Getting-Started.md

File metadata and controls

45 lines (39 loc) · 1.83 KB
Raw

Getting Started

Author: Pradyumna Krishna
UPDATED: 24/12/2020


The Structure* of this Repository is like:

Facial-AI
├── Docs
│   └── Docs related to this repository
├── FER
│   ├── edit.py
│   ├── haarcascade_frontalface_default.xml
│   ├── main.py
│   ├── model_weight.h5
│   ├── model.json
│   └── predict.py
├── templates
│   └── HTML templates that rendered by flasks
├── app.yaml
├── main.py
├── Procfile
└── requirements.txt
  • Docs: This Directory contains docs related to this repository.
  • FER: This Directory contains main files related to the web application. It contains a Flask Blueprint.
    • edit.py: This file take the image or file as an input and process it. Find no of faces and faces using haarcascade and apply TensorFlow model on them.
    • haarcascade_frontalface_default.xml: Used to detect faces.
    • main.py: Flask Blueprint, route web app, take user inputs and give output accordingly.
    • mode_weight.h5: Contains TensorFlow model used to predict emotions.
    • model.json: Contains layers of the TensorFlow model.
    • predict.py: Load the TensorFlow model and predict the emotion after taking input.
  • templates: This Directory Contains HTML Templates that are rendered by the flask.
  • app.yaml: Stores configuration data, GCP or Google App Engine Configuration for deploy this application.
  • main.py: Loads the Flask Blueprint present at FER directory and run it.
  • Procfile: Stores configuration data, Heroku configuration for deploy this application.
  • requirements.txt: Contains python libraries required for this web application to run.

Next: Setup

*Not including readmes or license files.