CoolFace
Apppublic

arasmodirpolimi/tech-stack-advisor-arasmo

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face
0likes
App README

🧠 Tech Stack Advisor – ML App (with Docker & Hugging Face Deployment)

Tech Stack Advisor is a hands-on machine learning project designed to teach you how to build, containerize, and deploy an ML-powered web application using Docker and Hugging Face Spaces.

🎯 This project is part of the "Artificial Intelligence and Machine Learning (AI/ML) with Docker" course from School of DevOps.

🚀 What You'll Learn

  • —Build and train a simple ML model using scikit-learn
  • —Create a UI using Gradio
  • —Containerize your app using a Dockerfile
  • —Push your Docker image to Docker Hub
  • —Deploy the Dockerized app on Hugging Face Spaces (free tier)

📁 Project Structure


tech-stack-advisor/
├── app.py             # Gradio web app
├── train.py           # Script to train and save ML model
├── requirements.txt   # Python dependencies
├── Dockerfile         # Docker build file (added during the lab)
├── model.pkl          # Trained ML model (generated after training)
├── encoders.pkl       # Encoders for categorical inputs (generated after training)
├── LICENSE            # Apache 2.0 license
└── README.md          # This guide

🧠 Step 1: Setup and Train Your ML Model

  1. 1.Clone the repository
bash
git clone https://github.com/<your-username>/tech-stack-advisor.git
cd tech-stack-advisor
  1. 1.Install dependencies

(Optional: Use a virtual environment)

bash
pip install -r requirements.txt
  1. 1.Train the model
bash
python train.py

This creates:

  • —model.pkl: the trained ML model
  • —encoders.pkl: label encoders for input/output features

🖥️ Step 2: Run the App Locally (Without Docker)

bash
python app.py

Visit the app in your browser at:

http://localhost:7860

🐳 Step 3: Add Docker Support

Create a file named Dockerfile in the root of the project:

dockerfile
FROM python:3.11-slim

WORKDIR /app

COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

COPY . .

EXPOSE 7860

CMD ["python", "app.py"]

🔧 Step 4: Build and Run the Docker Container

  1. 1.Build the image
bash
docker build -t tech-stack-advisor .
  1. 1.Run the container
bash
docker run -p 7860:7860 tech-stack-advisor

Visit: http://localhost:7860

☁️ Step 5: Publish to Docker Hub

  1. 1.Login to Docker Hub
bash
docker login
  1. 1.Tag the image
bash
docker tag tech-stack-advisor <your-dockerhub-username>/tech-stack-advisor:latest
  1. 1.Push it
bash
docker push <your-dockerhub-username>/tech-stack-advisor:latest

🌐 Step 6: Deploy to Hugging Face Spaces

  1. 1.Go to huggingface.co/spaces
  2. 2.Click Create New Space
  3. 3.Select:
  • —SDK: Docker
  • —Repository: Link to your GitHub repo with the Dockerfile
  • —Hugging Face will auto-build and deploy your container.

🧪 Test Your Skills

  • —Can you swap the model in train.py for a LogisticRegression model?
  • —Can you add logging to show which inputs were passed?
  • —Try changing the Gradio layout or theme!

🧾 License

This project is licensed under the Apache License 2.0. See the LICENSE file for details.

🛠 Happy shipping, DevOps and MLOps builders!