Metadata-Version: 2.1
Name: vsgan
Version: 1.0.0.post1
Summary: VapourSynth GAN Implementation using RRDBNet, based on ESRGAN's implementation
Home-page: https://gitlab.com/imPRAGMA/VSGAN
Author: PRAGMA
Author-email: pragma.exe@gmail.com
License: MIT
Description: # VSGAN
        VapourSynth GAN Implementation, based on ESRGAN's implementation
        
        # Installation
        `pip3 install vsgan`
        
        PyTorch and CUDA needs to be installed seperately.
        
        # Usage
        ```
        import vsgan as VSGAN
        (...)
        result = VSGAN.Start(clip=clip, model=[string: path to .pth], scale=[int: model's scale], old_arch=[bool: only set to true for models not working with the current arch])
        result.set_output()
        ```
        The resulting video will be the same color space format as your input clip, BUT, it's currently assuming its matrix is 709, if your inputs matrix is different, please change it [at this line of code](https://github.com/imPRAGMA/VSGAN/blob/master/vsgan.py#L55). :(
        
        Enjoy!
        Thanks to:
        - [xinntao](https://github.com/xinntao/ESRGAN) For RRDBNet_arch
        - [VideoHelp.com](https://videohelp.com) For helping me
        
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
