Skip to content

Development Builds

Jump to bottom
Benjamin Paine edited this page Jan 13, 2024 · 5 revisions

enfugue-0.3.3.dev1

Packaged 2024-01-13, includes DragNUWA and SVD pipeline improvements.

Environment Files

How to Use

  1. Select the file for your operating system, save someone on your computer as a .yml file.
  2. If you haven't already installed Conda, do so.
  3. Open a conda-initialized shell (bash, command-line, terminal, powershell, etc.) and run conda env create -f <downloaded_file>.yml. 3a. If you already have an enfugue environment, you may receive an error from this command. In this case, you should use conda env update -f <donwloaded_file.yml>.
  4. Run conda active enfugue followed by python -m enfugue run
  5. Navigate to the UI at https://app.enfugue.ai:45554 for local SSL or http://127.0.0.1:45555 for non-SSL/non-DNS/non-local connections.
Windows Environment File 
name: enfugue
channels:
  - pytorch
  - defaults
  - nvidia/label/cuda-11.8.0
  - conda-forge
dependencies:
  - cuda-toolkit=11.8.0
  - python=3.10.11
  - make=4.3
  - openh264
  - opencv
  - pip
  - numpy
  - 7zip
  - nodejs
  - ffmpeg
  - git
  - pip:
    - torch>=2.0.1+cu118
    - torchvision>=0.15.2+cu118
    - brotli>=1.0.9
    - cheroot>=9.0.0
    - nvidia-pyindex>=1.0.9
    - pibble[cherrypy]>=0.7.2
    - numpy>=1.24.3
    - colored>=1.4,<1.5
    - albumentations>=0.4.3,<0.5
    - pudb==2019.2
    - invisible-watermark>=0.2,<0.3
    - imageio>=2.31.1,<3.0
    - imageio-ffmpeg>=0.4.8,<0.5
    - pytorch-lightning>=2.0.6,<2.1
    - omegaconf>=2.1.1,<2.2
    - test-tube>=0.7.5,<0.8
    - streamlit>=0.73,<0.74
    - einops>=0.6.1,<0.7
    - torch-fidelity>=0.3,<0.4
    - transformers>=4.30,<4.33
    - torchmetrics>=1.1,<1.2
    - kornia>=0.6.10,<0.7
    - polygraphy>=0.47,<0.48
    - accelerate>=0.21,<0.22
    - tqdm>=4.27
    - safetensors>=0.3,<0.4
    - realesrgan>=0.3,<0.4
    - gfpgan>=1.3.8,<1.4
    - beautifulsoup4>=4.12,<5
    - pystray>=0.19,<0.24
    - pydantic==1.10.10
    - html2text==2020.1.16
    - polygraphy>=0.47,<0.48
    - mypy==1.2.0
    - mypy-extensions==1.0.0
    - types-requests>=2.30,<3.0
    - types-setuptools>=67.7,<68
    - types-urllib3>=1.26.25,<2.0
    - types-tabulate>=0.9,<0.10
    - types-pyyaml>=6.0,<7.0
    - importchecker>=2.0,<3.0
    - pyinstaller>=5.13.0,<6.0
    - pyarrow>=12.0.1,<13
    - torchsde>=0.2.5,<0.3
    - timm>=0.9.2,<1.0
    - lycoris-lora>=1.8
    - certifi>=2023.5.7
    - openmim>=0.3.9,<0.4
    - opensimplex>=0.4.5,<0.5
    - sentencepiece>=0.1
    - compel>=2.0
    - diffusers>=0.25.0
    - moviepy>=1.0
    - open-clip-torch>=2.20
    - xformers>=0.0.20
    - git+https://github.com/CompVis/taming-transformers.git@master#egg=taming-transformers
    - git+https://github.com/openai/CLIP.git@main#egg=clip
    - git+https://github.com/CompVis/stable-diffusion.git@main#egg=latent-diffusion
    - https://cdn.enfugue.ai/enfugue-0.3.3.dev1.tar.gz
    - --extra-index-url https://download.pytorch.org/whl/cu118
    - --extra-index-url https://pypi.ngc.nvidia.com
Linux Environment File 
name: enfugue
channels:
  - pytorch
  - defaults
  - nvidia/label/cuda-11.8.0
  - conda-forge
dependencies:
  - python=3.10.11
  - pip
  - cuda-toolkit=11.8.0
  - numpy
  - make=4.3
  - nodejs
  - ffmpeg
  - opencv=4.6.0
  - openh264
  - git
  - pip:
    - torch>=2.0.1+cu118
    - torchvision>=0.15.2+cu118
    - brotli>=1.0.9
    - cheroot>=9.0.0
    - nvidia-pyindex>=1.0.9
    - pibble[cherrypy]>=0.7.2
    - numpy>=1.24.3
    - colored>=1.4,<1.5
    - albumentations>=0.4.3,<0.5
    - pudb==2019.2
    - invisible-watermark>=0.2,<0.3
    - imageio>=2.31.1,<3.0
    - imageio-ffmpeg>=0.4.8,<0.5
    - pytorch-lightning>=2.0.6,<2.1
    - omegaconf>=2.1.1,<2.2
    - test-tube>=0.7.5,<0.8
    - streamlit>=0.73,<0.74
    - einops>=0.6.1,<0.7
    - torch-fidelity>=0.3,<0.4
    - transformers>=4.30,<5.0
    - torchmetrics>=1.1,<1.2
    - kornia>=0.6.10,<0.7
    - polygraphy>=0.47,<0.48
    - accelerate>=0.21,<0.22
    - tqdm>=4.27
    - safetensors>=0.3,<0.4
    - realesrgan>=0.3,<0.4
    - gfpgan>=1.3.8,<1.4
    - beautifulsoup4>=4.12,<5
    - pystray>=0.19,<0.24
    - pydantic==1.10.10
    - html2text==2020.1.16
    - mypy==1.2.0
    - mypy-extensions==1.0.0
    - types-requests>=2.30,<3.0
    - types-setuptools>=67.7,<68
    - types-urllib3>=1.26.25,<2.0
    - types-tabulate>=0.9,<0.10
    - types-pyyaml>=6.0,<7.0
    - importchecker>=2.0,<3.0
    - pyinstaller>=5.13.0,<6.0
    - pyarrow>=12.0.1,<13
    - torchsde>=0.2.5,<0.3
    - timm>=0.9.2,<1.0
    - certifi>=2023.5.7
    - lycoris-lora>=1.8
    - openmim>=0.3.9,<0.4
    - opensimplex>=0.4.5,<0.5
    - sentencepiece>=0.1
    - compel>=2.0
    - diffusers>=0.25.0
    - moviepy>=1.0
    - xformers>=0.0.20
    - open_clip_torch>=2.12.0
    - git+https://github.com/CompVis/taming-transformers.git@master#egg=taming-transformers
    - git+https://github.com/openai/CLIP.git@main#egg=clip
    - git+https://github.com/CompVis/stable-diffusion.git@main#egg=latent-diffusion
    - https://cdn.enfugue.ai/enfugue-0.3.3.dev1.tar.gz
    - --extra-index-url https://download.pytorch.org/whl/cu118
    - --extra-index-url https://pypi.ngc.nvidia.com

Motion Vector Controls

dragnuwa-controls.mp4
  • Click the Left-Mouse-Button on an empty portion of the canvas to start selecting existing points with a rectangular selector. Hold shift while doing this to add the selected points to your current selection, instead of replacing it. When you left-click on a point on the canvas, it will be grabbed and moved, optionally with shift as well to move all points at once. Left-clicking on a spline instead will select and move the entire spline.
  • Click Alt+Left-Mouse-Button on an empty section of the canvas to draw a new linear motion vector. Move your mouse and release it to draw a line between those points. When you use Alt and left-click on a point instead of the canvas, the point will be deleted. Alt-left-clicking a spline will add a new point along the segment your mouse is over.
  • Hold Ctrl+Shift and click the left-mouse-button anywhere on the canvas to rotate all selected points about their center.
  • When points are selected, press Delete on your keyboard to delete them.
  • When points are selected, press Ctrl+C on your keyboard to copy them. You must select at least two points on a spline for it to be copied.
  • Holding Ctrl while left-clicking will resume the previous behavior of moving the canvas.
Clone this wiki locally