Chris Ward

I had luck running on a 3090ti (24GB VRAM) with; ``` python gradio_demo.py --no_llava --use_image_slider --log_history --loading_half_params --use_tile_vae ``` but suddenly today I'm getting this fault

okay, this one was weird - I fired up `nvtop` and closed off some resources using VRAM and now I can run it again

This is an excellent repo - I've just seen obs firing up via x11 when running `make shell-dev`. I'm now wondering if it seems more reasonable to control scene selection...

``` CUDA_VISIBLE_DEVICES=0 python inference.py --checkpoint_path checkpoints/wav2lip_gan.pth --face samples/out.mp4 --audio samples/combined.wav Using cuda for inference. Reading video frames... Number of frames available for inference: 2770 (80, 8866) Length of mel chunks:...

same issue here - does not like :targets

thanks, but I don't see that in my interface? are you using an older version?

oh, it's a pop-out from the user button down the bottom left - thanks! just a UX issue then

ok, so the instructions are incorrect, and one of the shell scripts needs to be run... just not sure which one works or has been tested... ?

Duplicates Issue #78, outstanding since October 2020 Project is dead.

referencing another ticket that has been open for over a year; https://github.com/hirosystems/todos/issues/63