Hi there, I'm Henok 👋

I'm a Data Analyst and Python Programmer !

Technical Skills

Machine Learning API using FastAPI

Introduction

This project uses Docker to create a containerized application that serves a Hugging Face model with FastAPI. Docker is a tool that allows you to package and run applications in isolated environments, making them portable, secure, and easy to deploy. the docker file is on Hugging Face.

Hugging Face is a platform that provides state-of-the-art natural language processing models and datasets. FastAPI is a modern web framework that simplifies the development of RESTful APIs with Python. By combining these tools, you can build a scalable and robust machine learning app that can handle various natural language tasks.

Setup

For manual installation, you need to have Python3 on your system. Then you can clone this repo and follow the steps below:

• Windows:

    python -m venv venv; venv\Scripts\activate;
    python -m pip install -q --upgrade pip;
    python -m pip install -qr requirements.txt
  

• Linux & MacOs:

    python3 -m venv venv; source venv/bin/activate;
    python -m pip install -q --upgrade pip;
    python -m pip install -qr requirements.txt
  

The both long command-lines have a same structure, they pipe multiple commands using the symbol ; but you may manually execute them one after another.

1. Create the Python's virtual environment that isolates the required libraries of the project to avoid conflicts;
2. Activate the Python's virtual environment so that the Python kernel & libraries will be those of the isolated environment;
3. Upgrade Pip, the installed libraries/packages manager to have the up-to-date version that will work correctly;
4. Install the required libraries/packages listed in the requirements.txt file so that it will be allow to import them into the python's scripts and notebooks without any issue.

NB: For MacOs users, please install Xcode if you have an issue.

How to Run

The project consists two application. The first one accepts inputs from the user and return result with its confidentiality score. The second on accepts csv file which has input of various rows that going to be predicted. So the app read the file and retrieve records and predict for each rows and append the result with confidentiality score.

1. App For Filling Form

• Adjust the path

   cd Docker_ML\New folder\Docker_ML\app
  

• Then run the python file

  py main.py
  

• Then browse to

  http://127.0.0.1:8000/docs
  

2. App for Uploading file

• Adjust the path

   cd Docker_ML\New folder\Docker_ML\app_for_file
  

• Then run the python file

  py main.py
  

• Then browse to

  http://127.0.0.1:8000/docs
  

3. Docker in Hugging Face

• You can also run the app from hugging face docker image, browse to

    https://henok21-sepsis-prediction-in-docker.hf.space/docs
  

Screenshots

App	Result	App for file	Result for file

-->

🤝 Connect with me:

Further Link

MEDIUM