Ignore patterns below:

Directory structure:
└── heartwise-ai-deepcoro_clip/
    ├── README.md                      # Project documentation and overview
    ├──  .cursorrules                  # Cursor editor rules
    ├── generate_dataset.ipynb         # Notebook for dataset generation
    ├── plot_result_example.ipynb      # Notebook for visualizing results
    ├── pyproject.toml                 # Python project dependencies and settings
    ├── run_sweep.ipynb                # Notebook for hyperparameter sweeps
    ├── run_sweep.py                   # Script for hyperparameter sweeps
    ├── text_embeddings.pkl            # Cached text embeddings
    ├── uv.lock                        # Lock file for dependencies
    ├── .cursorrules                   # Cursor editor rules
    ├── .pre-commit-config.yaml        # Pre-commit hook configurations
    ├── config/                        # Configuration files directory
    │   ├── base_config.yaml           # Base configuration settings
    │   ├── base_config_bk.yaml        # Backup of base configuration
    │   ├── runner.sh                  # Runner script for training
    │   ├── sweep_config.yaml          # Hyperparameter sweep configuration
    │   └── sweep_config_3.yaml        # Additional sweep configuration
    ├── dataloaders/                   # Data loading modules
    │   ├── __init__.py               # Package initialization
    │   ├── multi_video_dataset.py    # Multiple video handling dataset
    │   ├── simple_text_dataset.py    # Text data loading
    │   ├── stats_dataset.py          # Statistics dataset
    │   └── video_dataset.py          # Video data loading
    ├── models/                        # Neural network models
    │   ├── text_encoder.py           # Text encoding model
    │   ├── video_aggregator.py       # Video feature aggregation
    │   └── video_encoder.py          # Video encoding model
    ├── projects/                      # Project-specific implementations
    │   ├── __init__.py               # Package initialization
    │   └── contrastive_pretraining.py # Contrastive learning implementation
    ├── runners/                       # Training runners
    │   ├── __init__.py               # Package initialization
    │   └── video_constrative_learning.py # Video contrastive learning runner
    ├── scripts/                       # Training and utility scripts
    │   ├── run_sweep.sh              # Script for running parameter sweeps
    │   ├── train.sh                  # Training script
    │   └── train_model_multi_gpu.py  # Multi-GPU training script
    └── utils/                         # Utility functions and helpers
        ├── config.py                  # Configuration handling
        ├── ddp.py                     # Distributed data parallel utilities
        ├── enums.py                   # Enumeration definitions
        ├── files_handler.py           # File operations handling
        ├── logging.py                 # Logging utilities
        ├── losses.py                  # Loss function implementations
        ├── metrics.py                 # Evaluation metrics
        ├── parser.py                  # Command line argument parsing
        ├── parser_typing.py           # Type definitions for parser
        ├── registry.py                # Model and component registry
        ├── schedulers.py              # Learning rate schedulers
        ├── seed.py                    # Random seed management
        ├── text_encoder.py            # Text encoding utilities
        └── video.py                   # Video processing utilities

### Rules
- Use the project structure to navigate the codebase.
- Use the README.md file to get an overview of the project.
- Use the pyproject.toml file to get an overview of the dependencies.
- Use the uv.lock file to get an overview of the dependencies.
- Use the config/ directory to get an overview of the configurations.
- Use the models/ directory to get an overview of the models.
- Use the scripts/ directory to get an overview of the scripts.

ALWAYS ACTIVATE THE VENV BEFORE RUNNING ANY COMAMND
```source .venv/bin/activate```

# Ignore patterns below:

# Python
__pycache__/
*.py[cod]
*$py.class
*.so
.Python
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
*.egg-info/
.installed.cfg
*.egg

# Virtual Environment
.env
.venv
env/
venv/
ENV/

# IDE
.idea/
.vscode/
*.swp
*.swo

# Project specific
wandb/
outputs/
*.ipynb_checkpoints/
lightning_logs/

# Data and models
*.pkl
*.pth
*.ckpt
*.pt

# Logs and databases
*.log
*.sqlite3

# OS specific
.DS_Store
.DS_Store?
._*
.Spotlight-V100
.Trashes
ehthumbs.db
Thumbs.db