Dwayne Robinson
Dwayne Robinson
If it helps, I have an: - [operator spreadsheet here](https://1drv.ms/x/c/ee82f5c6f06c7371/EXFzbPDG9YIggO5qRQUAAAABS4ZmNGhY9EOrkc2z2ekdKA?e=Rwp54b) (see first sheet "Operators Rearranged") comparing operators from multiple libraries (ONNX, WebNN, TOSA, MILR linalg, StableHLO, PyTorch prims, Apple MIL,...
> how to generate with AMD on Windows? Not yet. EasyDiffusion uses [SD Kit](https://github.com/easydiffusion/sdkit/) for generation, and SD Kit's readme states: > We're looking to add support for textual inversion...
> will be merging it to beta soon. Cool. 😎 > haven't figured out how to support LoRA with the AMD version, but might be worth releasing to beta anyway....
> How i can test this? Once merged to the beta branch, it can be enabled in Settings (easier than messing with config.yaml `update_branch` and typing `git` commands):  >...
@cmdr2 : Thanks for the links - I'll have to peruse the commit. I'm intimately familiar with DirectML, ONNX, and ONNX Runtime, given I've contributed hundreds of code changes to...
> guidance of what? for what? The "guidance scale" (or if you want the overly long term "classifier free guidance scale") guides the image in the direction away from the...
> without additional online help, like a tooltip ... > Guidance and CFG are both cryptonyms Yeah, extra information is prudent for both the even more confusing A1111 term (what...
> Aside from its apparent lack of JSON import/export for generation parameters making me disinterested :( > I also almost never run automatic1111 on my current machine And Easy Diffusion...
> So my question is, is that possible to calculate and add its output shape info to the graph during ORT Web graph optimization? Are you saying that ORT is...
/azp run Windows ARM64 QNN CI Pipeline,Windows x64 QNN CI Pipeline,Windows CPU CI Pipeline,Windows GPU CI Pipeline,Windows GPU TensorRT CI Pipeline,ONNX Runtime Web CI Pipeline,Linux CPU CI Pipeline,Linux CPU Minimal...