Method decode_video

FixModel.py:211–213

(self, latents, tiled=True, tile_size=(34, 34), tile_stride=(18, 16))

    
    
    def decode_video(self, latents, tiled=True, tile_size=(34, 34), tile_stride=(18, 16)):
        frames = self.vae.decode(latents, device=self.device, tiled=tiled, tile_size=tile_size, tile_stride=tile_stride)
        return frames


    @torch.no_grad()