Style Transfer VAE (CVAE) - Pytorch

This project implements a Style Transfer VAE (CVAE) using Pytorch, aimed at transforming images into specific painting genres from the "Best Artworks of all Time" dataset. The pipeline is structured with ZenML, and experiments are tracked using MLFlow.

Installation

Prerequisites

• Python 3.10

Install Pytorch and Related Packages

Use Conda to install Pytorch and related libraries:

conda install pytorch==2.2.2 torchvision==0.17.2 torchaudio==2.2.2 pytorch-cuda=11.8 -c pytorch -c nvidia

Install Other Dependencies

Install other required Python libraries:

pip install -r requirements.txt

Dataset Preparation

Download the dataset from Kaggle: Best Artworks of all Time

Dataset Structure

After downloading, extract the dataset into the data folder within this repository. Ensure the following structure:

data/
    ├── Images/         # Original images
    ├── resized/        # Resized images
    └── artists.csv     # Metadata file

Configuration and Setup

Initialize ZenML (First Time Only)

zenml init

Set Up MLFlow for Experiment Tracking

Install and configure MLFlow with ZenML: (First Time Only)

zenml integration install mlflow -y
zenml experiment-tracker register mlflow_tracker --flavor=mlflow
zenml model-deployer register mlflow --flavor=mlflow
zenml stack register mlflow_stack -a default -o default -d mlflow -e mlflow_tracker --set

Set the MLFlow stack:

zenml stack set mlflow_stack

Start ZenML:

• For Windows:

  zenml up --blocking

• For Linux/Mac:

  zenml up

Training and Evaluation

Run the training and evaluation pipeline:

python run_training_pipeline.py

After execution, a MLFlow tracking URI will be generated. To access the MLFlow UI:

mlflow ui --backend-store-uri {YOUR_TRACKING_URI_HERE}

Stop ZenML

To shut down ZenML gracefully, use the following command:

zenml down