Edit model card
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

Warning: based on original repo: https://git.nixnet.services/DUOLabs333/Photopea-Offline

Online-demo: https://fotopee-online.hf.space

How to use

  1. Download the repository: git clone https://huggingface.co/fotopee/fotopee
  2. Open a terminal: cd into the folder where you downloaded it: cd fotopee; in the folder, type: python run.py and press Enter (recommends: python run.py --update --fonts, see Notes); or just with path: python fotopee/run.py
  3. Open local link that was printed to the console (e.g.: http://localhost:7860/). It should load, and you can use it like normal.

Notes

  • If you want to update main photopea files, just append --update to the run.py file.
  • If you want to download the fonts, just append --fonts to the run.py file.
  • If you want to download the templates, just append --templates to the run.py file.
  • fotopee occupy about 7.6 Mb of space, fonts - 2.8 GB , and templates - 3.9 GB.
  • In order to add text to images, it is necessary to install fonts.

Deploy on HuggingFace Spaces (and similar things)

See demo

  1. Log-in to HF

  2. Create new Docker Space with Blank template

  3. In 'Files'-tab create new file with name Dockerfile (w/o extension)

  4. Paste code like this (and remove # comments):

    FROM python:3.12  # or python:3.11 or python:3.10
    WORKDIR /code
    RUN git clone https://huggingface.co/fotopee/fotopee  # replace repo url to yours
    COPY . .
    RUN chmod -R u+w /code
    RUN find /code -type d -exec chmod 0777 {} \;  # Suchlike permissions changes are required to run in HF.
    RUN find /code -type f -exec chmod 0666 {} \;
    EXPOSE 7860
    CMD ["python", "/code/fotopee/run.py", "--update", "--fonts", "--templates"]  # If you want, remove unneeded args.
    

    Save and wait until the building is complete, all assets have been downloaded, and the server is up.

    Note: Downloading fonts and templates can be a long process: many small files with a total size of over 6+ GB are downloaded! Luckily, this occurs only once.

Downloads last month
0
Unable to determine this model's library. Check the docs .