ZenML Alt Text Fine-Tuning Project

A machine learning pipeline built with ZenML for fine-tuning LLMs (default Gemma 3n) to generating alternative text descriptions for images.

📋 Prerequisites

• Python 3.12
• uv - Fast Python package installer and resolver
• ZenML - MLOps for Reliable AI
• unsloth - Easily finetune & train LLMs
• Hugging Face Account - Required for model access (token needed)
• OpenAI API Key - Required for OpenAI services

🛠️ Installation

1. Clone the repository:

git clone https://github.com/scheidti/zenml-alt-text.git
cd zenml-alt-text

2. Install uv (if not already installed)

3. Create a virtual environment and install dependencies:

uv venv
source .venv/bin/activate
uv sync

4. Set up environment variables:

cp .env.example .env
# Edit .env file with your configuration

5. Run pipelines:

# Data preparation pipeline
python run.py --pipeline data_preparation

# Batch processing pipeline
python run.py --pipeline batch_processing

# Training pipeline
python run.py --pipeline training

⚙️ Configuration

Each pipeline is configured through YAML files in the configs/ directory:

• data_preparation.yaml: Configuration for data preprocessing
• batch_processing.yaml: Configuration for batch inference
• training.yaml: Configuration for model training

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.