ML-Web-Integration

Web integration of ML projects

About

Website link: https://ml-integrations.onrender.com/

Warning: The website is hosted on a free service platform with highly limited resources below the recommendations of machine learning models. The website loading may be slow and the webcam streaming may hang. To fully experience the website, it is recommended to clone the repository and use local webhosting.

It is advised to not use the word classifier section (using PREDICT button), human emotions section and uploading image of more than 1 MB in the image caption section from the above link as it will crash the website (usage of memory above free quota).

The backend is written in Python based on Django. The website aims to feature most of the Machine Learning projects done by me and stack them in one place. build.sh is for production purpose and can be ignored.

The repositories of the projects included:

• Instagram Filters
• Word Classifier
• Human Emotions
• Image_Caption

This is project is now abondaned and no new AI projects will be added.

Installation

• Python 3.x version.
• requirements.txt for the dependencies.

Clone the repository

git clone https://github.com/SAM-DEV007/ML-Web-Integration.git
cd ML-Web_Integration

Create a virtual environment (optional)

python -m venv .venv
.venv\Scripts\activate.bat

Requirements

pip install -r requirements.txt

• In case, download_model.py files fails to download the model. Download the models and save it in the respective location:

  • ML-Web-Integration/word_classifier/Model/WordClassifier_Model.h5
  • ML-Web-Integration/human_emotions/Model/HumanEmotions_Model.h5

Initialize

python manage.py makemigrations
python manage.py migrate
python manage.py collectstatic

Run the local webserver

127.0.0.1 is for the local machine only. To apply for the available IP (use with other devices) - 0.0.0.0.

daphne -b 127.0.0.1 -p 8000 ML_Integration.asgi:application