08Aristodemus24 / micro-organism-classifier

This project aims to classify different micro-organisms using their respective microscopic images. Built with Svelte.js, Flask, and Tensorflow

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DEVELOPMENT FINISHED, DEPLOYMENT PENDING DUE TO EXCEEDING FILE SIZE OF 300MB

This project aims to classify different micro-organisms using their respective microscopic images. Built with React.js, Flask, and Tensorflow

requirements:

  1. git
  2. conda
  3. python

Source code usage

  1. assuming git is installed clone repository by running git clone https://github.com/08Aristodemus24/<repo name>
  2. assuming conda is also installed run conda create -n <environment name e.g. some-environment-name> python=3.11.5. Note python version should be 3.11.5 for the to be created conda environment to avoid dependency/package incompatibility.
  3. run conda activate <environment name used> or activate <environment name used>.
  4. run conda list -e to see list of installed packages. If pip is not yet installed run conda install pip, otherwise skip this step and move to step 5.
  5. navigate to directory containing the requirements.txt file.
  6. run pip install -r requirements.txt inside the directory containing the requirements.txt file
  7. after installing packages/dependencies run python index.py while in this directory to run app locally

App usage:

  1. control panel of app will have 1 input: The image field which allows the user to upload an image and then upload it to the server for further preprocessing and subsequently fed to the trained model to predict a probability which will further be preprocessed to translate it from probability to prediction

File structure:

|- client-side
    |- public
    |- src
        |- assets
            |- mediafiles
        |- boards
            |- *.png/jpg/jpeg/gig
        |- components
            |- *.svelte/jsx
        |- App.svelte/jsx
        |- index.css
        |- main.js
        |- vite-env.d.ts
    |- index.html
    |- package.json
    |- package-lock.json
    |- ...
|- server-side
    |- modelling
        |- data
        |- figures & images
            |- *.png/jpg/jpeg/gif
        |- final
            |- misc
            |- models
            |- weights
        |- metrics
            |- __init__.py
            |- custom.py
        |- models
            |- __init__.py
            |- arcs.py
        |- research papers & articles
            |- *.pdf
        |- saved
            |- misc
            |- models
            |- weights
        |- utilities
            |- __init__.py
            |- loaders.py
            |- preprocessors.py
            |- visualizers.py
        |- __init__.py
        |- experimentation.ipynb
        |- testing.ipynb
        |- training.ipynb
    |- static
        |- assets
            |- *.js
            |- *.css
        |- index.html
    |- index.py
    |- server.py
    |- requirements.txt
|- demo-video.mp4
|- .gitignore
|- readme.md

About

This project aims to classify different micro-organisms using their respective microscopic images. Built with Svelte.js, Flask, and Tensorflow


Languages

Language:Jupyter Notebook 97.4%Language:Python 1.5%Language:JavaScript 0.6%Language:CSS 0.5%Language:HTML 0.0%