swift-diffusion
swift-diffusion copied to clipboard
liuliu
I did a fresh clone and followed all instructions in the readme with M1 mac. I get ``` rhos_c 0.5, rhos_c 0.5 rhos_c 0.081319645, rhos_c 0.43422216 rhos_c 0.081311986, rhos_c 0.43350834...
INFO:python_coreml_stable_diffusion.chunk_mlprogram:Program loaded in 3.4 seconds INFO:python_coreml_stable_diffusion.chunk_mlprogram:Converting the two programs Running MIL frontend_milinternal pipeline: 0 passes [00:00, ? passes/s] Running MIL default pipeline: 100%|██████████████████████████████████████████████████| 66/66 [00:05
I was dong testing over multiple OS versions, but it seems to crash inside a UTM VM with M1 mac. Any setting with UTM i can do so that it...
manceShaders/MPSCore/ Types/MPSNDArray.mm:705: failed assertion '[MPSTemporaryNDArray initWithDevice:descriptor:] Error: product of dimension sizes > 2**31' This issue only happened once, and i am not able to reproduce it. Any idea what could...
when trying to compile and run through bazel according to your instruction on MacBook Pro M1 with Xcode 13.4 and swift 5.4, I was getting Float16 not available for macOS...
run now in the txt2img example you have one big graph.withNoGrad and the for loop is inside it. rather than that, if i have one graph.withNoGrad for text model stuff...
Has anyone tried int8 quantization ( maybe for only some layers) for SD and gotten good results. From what i know, no one has gotten good results yet.
Eta noise seed delta value will be added to seed and used to initialize RNG for noises when using samplers with Eta. (Especially useful in the Draw things App) See...
I have seen it very often the image has watermark like texture on the image. Sometime it's text like texture. Some other time the image is composed of several sub-images....