Benjamin Bossan

> WDYT about a non-exhaustive list of changes you envision would be needed to achieve [#1935 (comment)](https://github.com/huggingface/peft/issues/1935#issuecomment-2271539904). We could restrict that to LoKr, for example. Then we could have a...

> would be really helpful if we could have the docs here: Or if you could give us a pointer where the functional API is being used. I couldn't find...

_not stale_

Thanks @yaswanth19 for taking this up. As you correctly noted, this would be akin to adding a completely new PEFT method. However, the existing methods could be used to compare...

Yes, exactly @yaswanth19, it should leverage LyCORIS to do the heavy lifting of the logic of the respective method.

> I will try to refine the doc and put more example in this week, optimistically these 2 days. Thanks a lot. > I can see lycoris utils being used...

Thanks for providing the reproducer. When I tested it locally, the issue stemmed from gradient checkpointing. When setting `gradient_checkpointing=False`, there were gradients. Could you please give this a try? I...

Thanks for testing and confirming. Here is my slightly modified script: ```python import random import torch from torchvision.transforms import transforms from datasets import Dataset from qwen_vl_utils import process_vision_info from transformers...

Thanks for answering my questions, your answers make sense. > MPOP decomposes the parameter matrix W into multiple smaller tensors through a tensor-network representation (we use 5 in the paper)....

Thanks for bringing this to our attention @YuanheZ. I haven't checked the details, but please correct me if I misunderstand: For LoRA-One, we would also need to implement the technique...