Skip to content

windows_install.md

Latest commit

 

History

History
194 lines (144 loc) · 10.6 KB

windows_install.md

File metadata and controls

194 lines (144 loc) · 10.6 KB

Developer Installation (Windows)

This section provides instructions for installing the app (CLI and GUI) from source on Windows.

Note

This guide is for Windows. If you're using Linux or WSL, follow the Linux Installation.

Standalone .exe files are available here.

Install CLI

  1. Install system dependencies

    Open a PowerShell window as Administrator and run the following commands to install required programs via winget:

    winget install --id Gyan.FFmpeg -e --source winget
winget install --id Git.Git -e --source winget
winget install --id astral-sh.uv -e --source winget
set-ExecutionPolicy RemoteSigned

    Close the PowerShell window once the installation is complete.

  2. Get the source code

    Open a PowerShell window as a regular user. You will not need an Administrator Shell for any of the remaining steps.

    git clone https://codeberg.org/ladaapp/lada.git
cd lada

  3. Create a virtual environment to install python dependencies

    uv venv
.\.venv\Scripts\Activate.ps1

  4. Install Python dependencies

    Extra Supported GPU architectures Graphics Driver / Prerequisites Notes
    nvida-legacy Nvidia Maxwell(5.0), Pascal(6.0), Volta(7.0), Turing(7.5), Ampere(8.0, 8.6), Hopper(9.0) Nvidia GPU driver version >= 560 For RTX 10xx
    nvidia Nvidia Volta(7.0), Turing(7.5), Ampere(8.0, 8.6), Hopper(9.0), Blackwell(10.0, 12.0) Nvidia GPU driver version >= 570 For RTX 16xx, RTX 20xx up to including RTX 50xx
    intel Intel Discrete Arc GPUs: A-series (Alchemist), B-series (Battlemage)
    Intel Integrated Arc GPUs of Core Ultra Processors: Meteor Lake-H, Arrow Lake-H, Lunar Lake    		 Intel GPU Driver Installation docs
    cpu - Running Lada on CPU will be so slow that it's not really practical

    Based on your hardware, select the appropriate extra from the table above and install it with uv.

    You have to choose a single option in case your system contains GPUs of multiple vendors like an integrated Intel GPU and a dedicated Nvidia GPU.

    uv sync --extra nvidia # Adjust extra according to your available hardware

    Before continuing let's test if the installation was successful by checking if PyTorch detects your GPU (skip if using CPU):

    # Nvidia
uv run --no-project python -c "import torch ; print(torch.cuda.is_available())"
# Intel
uv run --no-project python -c "import torch ; print(torch.xpu.is_available())"

    If this prints True then you're good. If False, check your GPU drivers are up-to-date and ensure you've selected the right extra for your hardware.

  5. Apply patches

    uv pip install patch
uv run --no-project python -m patch -p1 -d .venv/lib/site-packages patches/increase_mms_time_limit.patch
uv run --no-project python -m patch -p1 -d .venv/lib/site-packages patches/remove_ultralytics_telemetry.patch
uv run --no-project python -m patch -p1 -d .venv/lib/site-packages patches/fix_loading_mmengine_weights_on_torch26_and_higher.diff
uv pip uninstall patch

  6. Download model weights

    Download the necessary model weights from HuggingFace

    Invoke-WebRequest 'https://huggingface.co/ladaapp/lada/resolve/main/lada_mosaic_detection_model_v2.pt?download=true' -OutFile ".\model_weights\lada_mosaic_detection_model_v2.pt"
Invoke-WebRequest 'https://huggingface.co/ladaapp/lada/resolve/main/lada_mosaic_detection_model_v4_accurate.pt?download=true' -OutFile ".\model_weights\lada_mosaic_detection_model_v4_accurate.pt"
Invoke-WebRequest 'https://huggingface.co/ladaapp/lada/resolve/main/lada_mosaic_detection_model_v4_fast.pt?download=true' -OutFile ".\model_weights\lada_mosaic_detection_model_v4_fast.pt"
Invoke-WebRequest 'https://huggingface.co/ladaapp/lada/resolve/main/lada_mosaic_restoration_model_generic_v1.2.pth?download=true' -OutFile ".\model_weights\lada_mosaic_restoration_model_generic_v1.2.pth"

    If you're interested in running DeepMosaics' restoration model you can also download their pretrained model clean_youknow_video.pth

    Invoke-WebRequest 'https://drive.usercontent.google.com/download?id=1ulct4RhRxQp1v5xwEmUH7xz7AK42Oqlw&export=download&confirm=t' -OutFile ".\model_weights\3rd_party\clean_youknow_video.pth"

You can now run the CLI with lada-cli.

Tip

Remember: To start Lada ensure you:

  • cd into the project root directory
  • Activate the virtual environment with .\.venv\Scripts\Activate.ps1
  • Run the CLI with lada-cli

Install GUI

  1. Install the CLI as per instructions above

  2. Install system dependencies

    You can choose between two methods to install system dependencies for the GUI:

    Option 1: Install pre-compiled dependencies (Recommended)

    • Download the pre-compiled system dependencies from here (sha256: 171152e8df65556f02065e080ec087153aaaa39634346b3dbe08a4f0f0d3ba1f).
    • Extract the file and ensure the build_gtk folder is located in the project root. The directory should look like this: 
      <lada root>
├── pyproject.toml
├── LICENSE.md
├── lada/
├── build_gtk/ # <- extracted from lada_windows_dependencies_python313_gvsbuild202611.7z
...

    Option 2: Build system dependencies yourself

    • Open a PowerShell window as Administrator and install the required build tools using winget:

      winget install --id MSYS2.MSYS2 -e --source winget
winget install --id Microsoft.VisualStudio.2022.BuildTools -e --source winget --silent --override "--wait --quiet --add ProductLang En-us --add Microsoft.VisualStudio.Workload.VCTools --includeRecommended"
winget install --id Rustlang.Rustup -e --source winget
winget install --id Microsoft.VCRedist.2013.x64  -e --source winget
winget install --id Microsoft.VCRedist.2013.x86  -e --source winget

      Then restart your computer.

    • Open a PowerShell as a regular user and prepare the build envirobnment (You will not need an Administrator Shell for any of the remaining steps)

      uv venv venv_gtk
.\venv_gtk\Scripts\Activate.ps1
uv pip install gvsbuild==2026.1.0
uv pip install patch
uv run --no-project python -m patch -p1 -d venv_gtk/lib/site-packages patches/gvsbuild_ffmpeg.patch
uv pip uninstall patch

    • Build the dependencies with gvsbuild. Grab a coffee, this will take a while...

      gvsbuild build --configuration=release --build-dir='./build_gtk' --enable-gi --py-wheel gtk4 adwaita-icon-theme pygobject libadwaita gstreamer gst-plugins-base gst-plugins-good gst-plugins-bad gst-plugins-ugly gst-plugin-gtk4 gst-libav gst-python gettext

    • Once the build is complete, deactivate the build environment and reactivate the project venv:

      deactivate
.\.venv\Scripts\Activate.ps1

  3. Install python dependencies

    Install the Python wheels we've built (or downloaded) in the previous step

    uv pip install --force-reinstall (Resolve-Path ".\build_gtk\gtk\x64\release\python\pygobject*.whl")
uv pip install --force-reinstall (Resolve-Path ".\build_gtk\gtk\x64\release\python\pycairo*.whl")

Tip

If you intend to hack on the GUI code install also the gui-dev group: uv pip install --group gui-dev.

You can now run the GUI with lada.

Tip

Remember: To start Lada ensure you:

  • cd into the project root directory
  • Activate the virtual environment with .\.venv\Scripts\Activate.ps1
  • Run the GUI with lada

Install Translations (optional)

If you prefer the app in a language other than English, you can use translation files if available.

  1. Install system dependencies

    We'll need the tool gettext. This program is already available if you've installed the GUI and you can continue with the next step.

    Alternatively, if you've only installed the CLI then it might be the easiest to install gettext this way:

    • Go to GNU gettext tools for Windows and download the latest release .zip file.
    • Extract the archive
    • Add the extracted directory to the $PATH environment variable: $env:Path = "<path/to/gettext/directory>" + ";" + $env:Path

  2. Compile translations

    powershell -ExecutionPolicy Bypass .\translations\compile_po.ps1

GUI and CLI should now show translations (if available) based on your system language settings (Time & language | Language & region | Windows display language).

Alternatively, you can set the environment variable LANGUAGE to your preferred language e.g. $env:LANGUAGE = "zh_CN". Using Windows settings is the preferred method though as just setting the environment variable may miss to set up the correct fonts.