stable_diffusion.openvino
stable_diffusion.openvino copied to clipboard
bes-dev
openvino still runs faster than diffusers-onnxruntime cpu, but changing size and IR format model conversion from onnx still is not understood. We can add DDIM support and produce good results...
I am looking to get the models in 16bit, currently I see its only 32bit.
The normal text-to-image works great but image-to-image fails when using an initial image with `python demo.py --prompt "Paint the dog red" --init-image ./dog.png` I guess it's related to the Openvino...
I think it would be useful for the average user to have two changes to the readme for a smoother install: 1) Cloning and cd'ing in the repo is missing...
I see that PR #98 has been merged but I don't see neither docker files, nor README.md lines I added...
Hi. Can you think of a way to infer between two images? I thought subtracting the latency between two images, but that doesn't seem to work because it doesn't infer...
I guessed with my limited knowledge: `parser.add_argument("--output", type=str, default=str(parser.parse_args()), help="output image name")` But it gives error: ``` line 45, in main cv2.imwrite(args.output, image) cv2.error: OpenCV(4.6.0) D:\a\opencv-python\opencv-python\opencv\modules\imgcodecs\src\loadsave.cpp:730: error: (-2:Unspecified error) could...
Hey, no idea how to do a discussion for this, but ... it would be great if this project could be updated to be more like the stable-diffusion-webui, with support...
I use the default settings, 32 inference steps, but after 18 iterations I already get my result which of course is poor because 18 iterations is not enough to get...
Default demo.py output file name changed from static output.png to timestamped filename a for better user experience. original: output.png new: out_YYYY-MM-DD_ss:mm:ss.png example: out_2022-10-23_23:46:53.png
