Installation

DeepAugment can be installed via pip or uv (recommended for faster installs).

Requirements

  • Python >= 3.11

  • PyTorch >= 2.0.0

  • CUDA-capable GPU (recommended for faster training)

Install from PyPI

Using pip:

pip install deepaugment

Using uv (faster):

uv add deepaugment

Install from Source

For development or the latest features:

git clone https://github.com/barisozmen/deepaugment.git
cd deepaugment
uv sync
uv pip install -e .

Verify Installation

Check that DeepAugment is properly installed:

import deepaugment
print(deepaugment.__version__)

You can also check available transforms:

from deepaugment import TRANSFORMS
print(f"Available transforms: {len(TRANSFORMS)}")

Dependencies

DeepAugment requires the following packages:

  • numpy >= 1.24.0 - Numerical operations

  • torch >= 2.0.0 - Deep learning framework

  • torchvision >= 0.15.0 - Image transformations

  • scikit-optimize >= 0.10.2 - Bayesian optimization

  • tqdm >= 4.66.0 - Progress bars

  • matplotlib >= 3.10.7 - Visualization

  • fire >= 0.6.0 - CLI interface

  • attrs >= 25.4.0 - Clean class definitions

  • cattrs >= 25.3.0 - Serialization

All dependencies are automatically installed with the package.

GPU Support

DeepAugment automatically detects and uses available GPUs. To verify GPU support:

import torch
print(f"CUDA available: {torch.cuda.is_available()}")
print(f"MPS available: {torch.backends.mps.is_available()}")

Device Selection

DeepAugment automatically selects the best available device by default:

from deepaugment import DeepAugment

# Auto-detect best device (default)
aug = DeepAugment(X_train, y_train, X_val, y_val, device="auto")

# Force specific device
aug = DeepAugment(X_train, y_train, X_val, y_val, device="cuda")  # NVIDIA GPU
aug = DeepAugment(X_train, y_train, X_val, y_val, device="mps")   # Apple Silicon
aug = DeepAugment(X_train, y_train, X_val, y_val, device="cpu")   # CPU only

Next Steps