StreamDiffusion
StreamDiffusion copied to clipboard
cumulo-autumn
Realtime-img2img
I installed it all. Running function ”demo | realtime-img2img“ with "python main.py --acceleration tensorrt" an error was thrown:
"The config attributes {'skip_prk_steps': True} were passed to LCMScheduler, but are not expected and will be ignored. Please verify your scheduler_config.json configuration file."
And a pop-up window prompts that python cannot find the entrance:
The camera unable to be called in web-UI and there returns the prompt "cannot read properties of undefined (reading 'enumerateDevices')" as I click the "START" buttom:
What did I do wrong? Everything else is normal
My environment:
python=3.10.9
CUDA=11.8
torch use:
pip3 install torch==2.1.0 torchvision==0.16.0 xformers --index-url https://download.pytorch.org/whl/cu118
tensorrt use:
pip install git+https://github.com/cumulo-autumn/StreamDiffusion.git@main#egg=streamdiffusion[tensorrt]
And make sure there is no problem with the "python -m streamdiffusion.tools.install-tensorrt" step.
win11x64
......
Please help me and thanx a lot~
The error message is still there, but I can make it work now: Enter "edge://flags/#unsafely-treat-insecure-origin-as-secure" in the edge address bar (in chrome just replace "edge" with "chrome"), and search with keywords "Insecure origins treated as secure", then click the "enabled" button. Add addresses in the box below: "http://localhost:7860,http://0.0.0.0:7860" and restart edge.
The error message is still there, but I can make it work now: Enter "edge://flags/#unsafely-treat-insecure-origin-as-secure" in the edge address bar (in chrome just replace "edge" with "chrome"), and search with keywords "Insecure origins treated as secure", then click the "enabled" button. Add addresses in the box below: "http://localhost:7860,http://0.0.0.0:7860" and restart edge.
This worked for me on chrome so thank you!
You can also just use localhost:7860 and then you need not make any settings changes.
