Update¶

User Setup¶

Option 1: Using uv (Recommended for Performance)¶

Activate your environment:

source qualia_env/bin/activate  # Unix/macOS
qualia_env\Scripts\activate     # Windows

Update Qualia and its dependencies:

# Basic update with PyTorch support
uv pip install -U qualia-core[pytorch]

# With additional dependencies
uv pip install -U "qualia-core[pytorch,visualize,test]"

# Update every package in the environnement
uv pip freeze | cut -d= -f1 | xargs uv pip install -U

Option 2: Using PDM (Recommended for Dependency Management)¶

Activate your environment:

$(pdm venv activate qualia_env)

Update Qualia and all dependencies:

# Update all packages in the environment
pdm update --update-all

# Update specific package with dependencies
pdm update qualia-core

Option 3: Using pip¶

Pip will only update the specified Qualia components and install new dependencies when required.

# Basic update with PyTorch support
pip install -U qualia-core[pytorch]

# With additional dependencies
pip install -U "qualia-core[pytorch,visualize,test]"

Developer Setup¶

Update Qualia’s Components¶

For developer setups, first update the source code:

Update the source code from the git repositories:

# Activate environment
source qualia_env/bin/activate  # Unix/macOS
qualia_env\Scripts\activate     # Windows

# Update repositories
cd qualia
git pull

Repeat for any additional Qualia components you’re using.

Update Third-party Dependencies¶

Option 1: Using uv¶

cd qualia
source qualia_env/bin/activate

# Update all dependencies
uv pip freeze | cut -d= -f1 | xargs uv pip install -U

# Development installation
cd qualia
uv pip install -U -e ./qualia-core[pytorch,clearml]

# Component-specific
cd qualia-core
uv pip install -U -e .[pytorch,clearml]

Option 2: Using PDM¶

cd qualia
$(pdm venv activate)

# Update all dependencies
pdm update --update-all

# Update specific component and its dependencies
pdm update qualia-core

Option 3: Using pip¶

Note: pip only updates dependencies when explicitly required.

Update the development installation:

pip install -U -e .[pytorch]

Repeat for additional components, following the dependency graph order (see Components).