MLOPs-and-LLMOPs-Demo

A mini demo for demonstrate the basic steps of MLOPs using public tools and platforms, including dataset management, model training, CI/CD

Prerequisites

Before starting with this project, ensure you have the following installed and configured:

1. Python (version 3.6 or higher)

   • Download and install Python from python.org.
   • Verify the installation by running:

     python --version

2. Git

3. VSCode(recommended)

4. Sign up for Hugging Face Hub Account

   • Sign up at huggingface.co

5. GitHub Account

---

Guidance

1. Fork and Clone this repo to your local machine

git clone [email protected]:Whynotmetoo/MLOPs-and-LLMOPs-Demo.git

2. Create virtual environment

python -m venv venv

3. Activate your virtual environment

• window:

venv\Scripts\activate

• Linux/macOS:

source venv/bin/activate

4. Install packages

pip install -r requirements.txt

5. Run the code to train model

python train_model.py

6. Configure CI/CD

1. Create Hugging Face Repo and access token
   • Creat a new model repo and copy the repo id on the top left corner of the repo, in a format as your_username/your_model_repo
   • Go to Settings > Access Tokens to generate your token.
2. Create github repository secrets
   • Save the token as a GitHub Secret named HF_TOKEN and the repo id named REPO_ID in your GitHub repository for secure access during deployment. Go the setting > Actions secrets and variables > Actions to add your Repository secrets

7. push you code

push you code to github and check the pipeline test