Peter Lin

Results 7 comments of Peter Lin

[BUG] Config mesh_device None

Got the same error here :( Any updates on this? Thanks!

Image Editting via Inversion

> > > Very cool examples. Is it possible to add these inverse feature into our code base? [@nttung1110](https://github.com/nttung1110) > > > > > > Sure, please let me know...

ONNX Feature already worked. This is a result run on GPU with onnxruntime. I'll preview and plan to new feature updated is TensorRT and C++ Inference code in the next week 🔥🔥🔥

@aihacker111 Thanks for your great work!! I would like to know where can I find the C++ inference code. :))

ONNX Feature already worked. This is a result run on GPU with onnxruntime. I'll preview and plan to new feature updated is TensorRT and C++ Inference code in the next week 🔥🔥🔥

> @ChunChenLin I'm currently still not implement C++ code , maybe in the next week :) @aihacker111 Sure, thanks! Looking forward to it. Thanks again for your great work. :))

ONNX Feature already worked. This is a result run on GPU with onnxruntime. I'll preview and plan to new feature updated is TensorRT and C++ Inference code in the next week 🔥🔥🔥

@aihacker111 Hey, how's everything :) Just wanna know is the C++ inference code available or coming soon? :))

No Significant Improvement Observed in Model Training Speed

@lianghsun Hi, thanks for your sharing! I also oberserved that there's no significant difference with/without Liger-Kernel on LLaMA-Factory. Are you able to get the benefit of Liger-Kernel yet, and under...

No Significant Improvement Observed in Model Training Speed

> Hi @ChunChenLin , may I ask which model you are using and whether you are utilizing any distributed training? @lianghsun Thanks for reply! I'm using `meta-llama/llama3-8B-instruct` model and utilizing...