https://github.com/bhimrazy/image-recognition-app-using-fastapi-and-pytorch

This is an image recognition application based on the FastAPI framework and PyTorch which uses pretrained DenseNet 121 model to detect the image.
https://github.com/bhimrazy/image-recognition-app-using-fastapi-and-pytorch

densenet-pytorch fastapi jinja2 python pytorch tailwindcss

Last synced: 3 months ago
JSON representation

This is an image recognition application based on the FastAPI framework and PyTorch which uses pretrained DenseNet 121 model to detect the image.

• Host: GitHub
• URL: https://github.com/bhimrazy/image-recognition-app-using-fastapi-and-pytorch
• Owner: bhimrazy
• License: mit
• Created: 2021-10-25T11:26:09.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2023-10-15T13:12:35.000Z (over 1 year ago)
• Last Synced: 2024-05-14T00:09:07.048Z (9 months ago)
• Topics: densenet-pytorch, fastapi, jinja2, python, pytorch, tailwindcss
• Language: Jupyter Notebook
• Homepage:
• Size: 6.54 MB
• Stars: 21
• Watchers: 3
• Forks: 4
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        


  

  


  Image Recognition App using FastAPI and PyTorch


In this project, I have tried to build and deploy an Image Recognition App using FastAPI and

PyTorch.



## Tutorial 

[![fastapi pytorch image recognition](https://user-images.githubusercontent.com/46085301/143211855-fe73e3b0-fc8c-4aa4-8428-e93cc16d6dfd.png)](https://youtu.be/7Rm5mGhYWVI)

## Installation

Run my Project

```bash

  git clone https://github.com/bhimrazy/Image-Recognition-App-using-FastAPI-and-PyTorch

  cd Image-Recognition-App-using-FastAPI-and-PyTorch

  python -m venv venv

  source venv/bin/activate

  pip install -r requirements.txt

  uvicorn main:app --reload

```

## Image Recognition App using FastAPI and PyTorch: TODO

- [x] Create a virtual environment

- [x] Create FastAPI App

    - Install fast API

    - Install Uvicorn

    - Install Pytest

    - Install Jinja2

    - Install python-multipart

    - Install requests

    - Create a main file with some routes

    - Create a main test file to test the home page route

- [x] Pytorch Setup

    - Install torch & torchvision (use cpu version for small size)

- [x] Prediction 

    - Create a predict post route

    - Create a file utils.py

    - Test predict route

    - Create some helper function

    - Put some test images inside static folder

    - Create a test to upload an image in predict route

    - Predict

- [x] Create a home page for prediction

    - Create an index.html file inside the templates directory

    - Setup template and static directory in the main app

    - Initial HTML for home page

    - Use Tailwind CSS cdn link for css

    - Google Fonts

    - Create a form to predict

    - update homepage route for prediction

    - Update UI of the page

    - Add Some javascript to autoload the image

    - Add logo and favicon

    - Add meta tags

    - Add response image for preview as base64 data

## 📚 RESOURCES:

◆ PyTorch: https://pytorch.org/tutorials/intermediate/flask_rest_api_tutorial.html 


◆ FastAPI: https://fastapi.tiangolo.com

## Author

- [@bhimrazy](https://www.github.com/bhimrazy)